
# Imports librairies
import sys
import warnings

import IPython as ip
import matplotlib.pyplot as plt
import missingno as msno
import sklearn as nltk
from sklearn.preprocessing import StandardScaler
from sklearn.preprocessing import PowerTransformer
# import mytools as tools
import numpy as np
import pandas as pd
import seaborn as sns
from IPython.display import display
import geopandas as gpd

warnings.filterwarnings("ignore")

# Versions
print("Les version des librairies : ")
print("Python       : " + sys.version)
print("Numpy        : " + np.version.full_version)
print("Pandas       : " + pd.__version__)
print("Seaborn      : " + sns.__version__)
print("missingno    : " + msno.__version__)
print("scikit-learn : " + nltk.__version__)

Les version des librairies : 
Python       : 3.10.7 (tags/v3.10.7:6cc6b13, Sep  5 2022, 14:08:36) [MSC v.1933 64 bit (AMD64)]
Numpy        : 1.23.4
Pandas       : 1.5.1
Seaborn      : 0.12.1
missingno    : 0.5.1
scikit-learn : 1.1.3


from jyquickhelper import add_notebook_menu

add_notebook_menu()


# On crée une fonction info_data pour les autres df (DRY)
def info_data(df):
    print(
        "------------------------------------------------------------------------------------------------------"
    )
    print(f"LES DONNES \n")
    print("Nombre de lignes :", df.shape[0], "lignes")
    print("Nombre de colonnes :", df.shape[1], "colonnes")
    # Recherche de doublon
    print(
        f"Nombre d'éléments dupliqués dans le dataframe : {country.duplicated().sum()} eléments"
    )
    nb_na = df.isnull().sum().sum()
    if nb_na > 0:
        nb_data_tt = np.product(df.shape)
        pourcent_na_tt = round((nb_na / nb_data_tt) * 100, 2)
        print(
            f"Nombre total de données manquantes dans le dataframe : {nb_na} données manquantes sur {nb_data_tt} ({pourcent_na_tt}%)"
        )
    else:
        print("Aucune valeur manquante")


# On crée une fonction get_na (DRY)
def info_data_na(df):
    nb_na = df.isnull().sum().sum()
    nb_data_tt = np.product(df.shape)
    pourcent_na_tt = round((nb_na / nb_data_tt) * 100, 2)
    print(
        f"Nombre total de données manquantes dans le dataframe : {nb_na} données manquantes sur {nb_data_tt} ({pourcent_na_tt}%)"
    )
    print(
        "------------------------------------------------------------------------------------------------------"
    )
    print("Visualisation des données manquantes")
    if nb_na > 0:
        plt.figure(figsize=(20, 10))
        plt.title("Visualisation des valeurs manquantes")
        sns.heatmap(df.isna(), cbar=False)
        plt.show()


# Import données
country = pd.read_csv("assets/data/EdStatsCountry.csv", sep=",")


# afficher les 5 premières et les 5 dernières lignes du dataframe
display(country)


# Résumé des variables qualitatives
country.describe(exclude=[np.number])


# Résumé des variables quantitatives
country.describe()


# On regarde les types des variables
country.dtypes

Country Code                                          object
Short Name                                            object
Table Name                                            object
Long Name                                             object
2-alpha code                                          object
Currency Unit                                         object
Special Notes                                         object
Region                                                object
Income Group                                          object
WB-2 code                                             object
National accounts base year                           object
National accounts reference year                     float64
SNA price valuation                                   object
Lending category                                      object
Other groups                                          object
System of National Accounts                           object
Alternative conversion factor                         object
PPP survey year                                       object
Balance of Payments Manual in use                     object
External debt Reporting status                        object
System of trade                                       object
Government Accounting concept                         object
IMF data dissemination standard                       object
Latest population census                              object
Latest household survey                               object
Source of most recent Income and expenditure data     object
Vital registration complete                           object
Latest agricultural census                            object
Latest industrial data                               float64
Latest trade data                                    float64
Latest water withdrawal data                          object
Unnamed: 31                                          float64
dtype: object


# Premières informations sur les données du dataframe
info_data(country)

------------------------------------------------------------------------------------------------------
LES DONNES 

Nombre de lignes : 241 lignes
Nombre de colonnes : 32 colonnes
Nombre d'éléments dupliqués dans le dataframe : 0 eléments
Nombre total de données manquantes dans le dataframe : 2354 données manquantes sur 7712 (30.52%)


# Lister les noms de colonne
country.columns

Index(['Country Code', 'Short Name', 'Table Name', 'Long Name', '2-alpha code',
       'Currency Unit', 'Special Notes', 'Region', 'Income Group', 'WB-2 code',
       'National accounts base year', 'National accounts reference year',
       'SNA price valuation', 'Lending category', 'Other groups',
       'System of National Accounts', 'Alternative conversion factor',
       'PPP survey year', 'Balance of Payments Manual in use',
       'External debt Reporting status', 'System of trade',
       'Government Accounting concept', 'IMF data dissemination standard',
       'Latest population census', 'Latest household survey',
       'Source of most recent Income and expenditure data',
       'Vital registration complete', 'Latest agricultural census',
       'Latest industrial data', 'Latest trade data',
       'Latest water withdrawal data', 'Unnamed: 31'],
      dtype='object')


# Nombre de valeurs manquantes par variable en %
round(country.isnull().mean() * 100, 2)

Country Code                                           0.00
Short Name                                             0.00
Table Name                                             0.00
Long Name                                              0.00
2-alpha code                                           1.24
Currency Unit                                         10.79
Special Notes                                         39.83
Region                                                11.20
Income Group                                          11.20
WB-2 code                                              0.41
National accounts base year                           14.94
National accounts reference year                      86.72
SNA price valuation                                   18.26
Lending category                                      40.25
Other groups                                          75.93
System of National Accounts                           10.79
Alternative conversion factor                         80.50
PPP survey year                                       39.83
Balance of Payments Manual in use                     24.90
External debt Reporting status                        48.55
System of trade                                       17.01
Government Accounting concept                         33.20
IMF data dissemination standard                       24.90
Latest population census                              11.62
Latest household survey                               41.49
Source of most recent Income and expenditure data     33.61
Vital registration complete                           53.94
Latest agricultural census                            41.08
Latest industrial data                                55.60
Latest trade data                                     23.24
Latest water withdrawal data                          25.73
Unnamed: 31                                          100.00
dtype: float64


# Information et visualisation des données manquantes
info_data_na(country)

Nombre total de données manquantes dans le dataframe : 2354 données manquantes sur 7712 (30.52%)
------------------------------------------------------------------------------------------------------
Visualisation des données manquantes


# Identifier les différentes manières de nommer les pays + Regfion et revenu
country[["Country Code", "Short Name","Table Name","Long Name","2-alpha code","WB-2 code" ,"Currency Unit","Region", "Income Group"]]


# Lister les regions
print("Régions :\n", country["Region"].unique())

Régions :
 ['Latin America & Caribbean' 'South Asia' 'Sub-Saharan Africa'
 'Europe & Central Asia' nan 'Middle East & North Africa'
 'East Asia & Pacific' 'North America']


print(country["Region"].nunique(), "regions")

7 regions


# compter et Lister les pays
print("Pays :\n", country["Short Name"].nunique(), "pays")
print(
    "Le fichier présente 241 données géographiques comprennant des pays et dses zones géographiques"
)

Pays :
 241 pays
Le fichier présente 241 données géographiques comprennant des pays et dses zones géographiques


# Dénominations pays
country[
    [
        "Country Code",
        "Short Name",
        "Table Name",
        "Long Name",
        "2-alpha code",
        "WB-2 code",
    ]
]


# https://data.worldbank.org/country
# On liste les zones contenu dans le df
areas = [
    "Arab World",
    "East Asia & Pacific",
    "East Asia & Pacific (excluding high income)",
    "East Asia & Pacific (developing only)",
    "East Asia & Pacific (all income levels)",
    "Euro area",
    "European Union",
    "Europe & Centraµl Asia",
    "Europe & Central Asia (excluding high income)",
    "Europe & Central Asia (developing only)",
    "Europe & Central Asia (all income levels)",
    "European Union",
    "Heavily indebted poor countries (HIPC)",
    "High income",
    "Latin America & Caribbean",
    "Latin America & Caribbean (developing only)",
    "Latin America & Caribbean (excluding high income)",
    "Least developed countries: UN classification",
    "Low & middle income",
    "Low income",
    "Lower middle income",
    "Middle East & North Africa",
    "Middle East & North Africa (excluding high income)",
    "Middle East & North Africa (all income levels)",
    "Middle East & North Africa (developing only)",
    "Middle income",
    "North America",
    "OECD members",
    "South Asia",
    "Sub-Saharan Africa",
    "Sub-Saharan Africa (excluding high income)",
    "Upper middle income",
    "Sub-Saharan Africa (developing only)",
    "Sub-Saharan Africa (all income levels)",
    "World",
]


# Compter le nombre de pays qui ne sont pas des zones géographiques
country[~country["Short Name"].isin(areas)]["Short Name"].unique().shape

(217,)


# Créer la liste des pays présent dans country
list_countries = (
    country[~country["Short Name"].isin(areas)]["Short Name"].unique().tolist()
)
print(list_countries)

['Aruba', 'Afghanistan', 'Angola', 'Albania', 'Andorra', 'United Arab Emirates', 'Argentina', 'Armenia', 'American Samoa', 'Antigua and Barbuda', 'Australia', 'Austria', 'Azerbaijan', 'Burundi', 'Belgium', 'Benin', 'Burkina Faso', 'Bangladesh', 'Bulgaria', 'Bahrain', 'The Bahamas', 'Bosnia and Herzegovina', 'Belarus', 'Belize', 'Bermuda', 'Bolivia', 'Brazil', 'Barbados', 'Brunei', 'Bhutan', 'Botswana', 'Central African Republic', 'Canada', 'Switzerland', 'Channel Islands', 'Chile', 'China', "Côte d'Ivoire", 'Cameroon', 'Dem. Rep. Congo', 'Congo', 'Colombia', 'Comoros', 'Cabo Verde', 'Costa Rica', 'Cuba', 'Curaçao', 'Cayman Islands', 'Cyprus', 'Czech Republic', 'Germany', 'Djibouti', 'Dominica', 'Denmark', 'Dominican Republic', 'Algeria', 'Ecuador', 'Egypt', 'Eritrea', 'Spain', 'Estonia', 'Ethiopia', 'Finland', 'Fiji', 'France', 'Faeroe Islands', 'Micronesia', 'Gabon', 'United Kingdom', 'Georgia', 'Ghana', 'Gibraltar', 'Guinea', 'The Gambia', 'Guinea-Bissau', 'Equatorial Guinea', 'Greece', 'Grenada', 'Greenland', 'Guatemala', 'Guam', 'Guyana', 'Hong Kong SAR, China', 'Honduras', 'Croatia', 'Haiti', 'Hungary', 'Indonesia', 'Isle of Man', 'India', 'Ireland', 'Iran', 'Iraq', 'Iceland', 'Israel', 'Italy', 'Jamaica', 'Jordan', 'Japan', 'Kazakhstan', 'Kenya', 'Kyrgyz Republic', 'Cambodia', 'Kiribati', 'St. Kitts and Nevis', 'Korea', 'Kuwait', 'Lao PDR', 'Lebanon', 'Liberia', 'Libya', 'St. Lucia', 'Latin America & Caribbean (all income levels)', 'Liechtenstein', 'Sri Lanka', 'Lesotho', 'Lithuania', 'Luxembourg', 'Latvia', 'Macao SAR, China', 'St. Martin (French part)', 'Morocco', 'Monaco', 'Moldova', 'Madagascar', 'Maldives', 'Mexico', 'Marshall Islands', 'Macedonia', 'Mali', 'Malta', 'Myanmar', 'Montenegro', 'Mongolia', 'Northern Mariana Islands', 'Mozambique', 'Mauritania', 'Mauritius', 'Malawi', 'Malaysia', 'Namibia', 'New Caledonia', 'Niger', 'Nigeria', 'Nicaragua', 'Netherlands', 'Norway', 'Nepal', 'Nauru', 'New Zealand', 'Oman', 'Pakistan', 'Panama', 'Peru', 'Philippines', 'Palau', 'Papua New Guinea', 'Poland', 'Puerto Rico', "Dem. People's Rep. Korea", 'Portugal', 'Paraguay', 'West Bank and Gaza', 'French Polynesia', 'Qatar', 'Romania', 'Russia', 'Rwanda', 'Saudi Arabia', 'Sudan', 'Senegal', 'Singapore', 'Solomon Islands', 'Sierra Leone', 'El Salvador', 'San Marino', 'Somalia', 'Serbia', 'South Sudan', 'São Tomé and Principe', 'Suriname', 'Slovak Republic', 'Slovenia', 'Sweden', 'Swaziland', 'Sint Maarten (Dutch part)', 'Seychelles', 'Syrian Arab Republic', 'Turks and Caicos Islands', 'Chad', 'Togo', 'Thailand', 'Tajikistan', 'Turkmenistan', 'Timor-Leste', 'Tonga', 'Trinidad and Tobago', 'Tunisia', 'Turkey', 'Tuvalu', 'Tanzania', 'Uganda', 'Ukraine', 'Uruguay', 'United States', 'Uzbekistan', 'St. Vincent and the Grenadines', 'Venezuela', 'Virgin Islands', 'Vietnam', 'Vanuatu', 'Samoa', 'Kosovo', 'Yemen', 'South Africa', 'Zambia', 'Zimbabwe']


# Compter le nombre d'éléments dans la liste pour avoir le nombre de pays
len(list_countries)

217


print(country["Short Name"].unique())

['Aruba' 'Afghanistan' 'Angola' 'Albania' 'Andorra' 'Arab World'
 'United Arab Emirates' 'Argentina' 'Armenia' 'American Samoa'
 'Antigua and Barbuda' 'Australia' 'Austria' 'Azerbaijan' 'Burundi'
 'Belgium' 'Benin' 'Burkina Faso' 'Bangladesh' 'Bulgaria' 'Bahrain'
 'The Bahamas' 'Bosnia and Herzegovina' 'Belarus' 'Belize' 'Bermuda'
 'Bolivia' 'Brazil' 'Barbados' 'Brunei' 'Bhutan' 'Botswana'
 'Central African Republic' 'Canada' 'Switzerland' 'Channel Islands'
 'Chile' 'China' "Côte d'Ivoire" 'Cameroon' 'Dem. Rep. Congo' 'Congo'
 'Colombia' 'Comoros' 'Cabo Verde' 'Costa Rica' 'Cuba' 'Curaçao'
 'Cayman Islands' 'Cyprus' 'Czech Republic' 'Germany' 'Djibouti'
 'Dominica' 'Denmark' 'Dominican Republic' 'Algeria'
 'East Asia & Pacific (developing only)'
 'East Asia & Pacific (all income levels)'
 'Europe & Central Asia (developing only)'
 'Europe & Central Asia (all income levels)' 'Ecuador' 'Egypt' 'Euro area'
 'Eritrea' 'Spain' 'Estonia' 'Ethiopia' 'European Union' 'Finland' 'Fiji'
 'France' 'Faeroe Islands' 'Micronesia' 'Gabon' 'United Kingdom' 'Georgia'
 'Ghana' 'Gibraltar' 'Guinea' 'The Gambia' 'Guinea-Bissau'
 'Equatorial Guinea' 'Greece' 'Grenada' 'Greenland' 'Guatemala' 'Guam'
 'Guyana' 'High income' 'Hong Kong SAR, China' 'Honduras'
 'Heavily indebted poor countries (HIPC)' 'Croatia' 'Haiti' 'Hungary'
 'Indonesia' 'Isle of Man' 'India' 'Ireland' 'Iran' 'Iraq' 'Iceland'
 'Israel' 'Italy' 'Jamaica' 'Jordan' 'Japan' 'Kazakhstan' 'Kenya'
 'Kyrgyz Republic' 'Cambodia' 'Kiribati' 'St. Kitts and Nevis' 'Korea'
 'Kuwait' 'Latin America & Caribbean (developing only)' 'Lao PDR'
 'Lebanon' 'Liberia' 'Libya' 'St. Lucia'
 'Latin America & Caribbean (all income levels)'
 'Least developed countries: UN classification' 'Low income'
 'Liechtenstein' 'Sri Lanka' 'Lower middle income' 'Low & middle income'
 'Lesotho' 'Lithuania' 'Luxembourg' 'Latvia' 'Macao SAR, China'
 'St. Martin (French part)' 'Morocco' 'Monaco' 'Moldova' 'Madagascar'
 'Maldives' 'Middle East & North Africa (all income levels)' 'Mexico'
 'Marshall Islands' 'Middle income' 'Macedonia' 'Mali' 'Malta' 'Myanmar'
 'Middle East & North Africa (developing only)' 'Montenegro' 'Mongolia'
 'Northern Mariana Islands' 'Mozambique' 'Mauritania' 'Mauritius' 'Malawi'
 'Malaysia' 'North America' 'Namibia' 'New Caledonia' 'Niger' 'Nigeria'
 'Nicaragua' 'Netherlands' 'Norway' 'Nepal' 'Nauru' 'New Zealand'
 'OECD members' 'Oman' 'Pakistan' 'Panama' 'Peru' 'Philippines' 'Palau'
 'Papua New Guinea' 'Poland' 'Puerto Rico' "Dem. People's Rep. Korea"
 'Portugal' 'Paraguay' 'West Bank and Gaza' 'French Polynesia' 'Qatar'
 'Romania' 'Russia' 'Rwanda' 'South Asia' 'Saudi Arabia' 'Sudan' 'Senegal'
 'Singapore' 'Solomon Islands' 'Sierra Leone' 'El Salvador' 'San Marino'
 'Somalia' 'Serbia' 'Sub-Saharan Africa (developing only)' 'South Sudan'
 'Sub-Saharan Africa (all income levels)' 'São Tomé and Principe'
 'Suriname' 'Slovak Republic' 'Slovenia' 'Sweden' 'Swaziland'
 'Sint Maarten (Dutch part)' 'Seychelles' 'Syrian Arab Republic'
 'Turks and Caicos Islands' 'Chad' 'Togo' 'Thailand' 'Tajikistan'
 'Turkmenistan' 'Timor-Leste' 'Tonga' 'Trinidad and Tobago' 'Tunisia'
 'Turkey' 'Tuvalu' 'Tanzania' 'Uganda' 'Ukraine' 'Upper middle income'
 'Uruguay' 'United States' 'Uzbekistan' 'St. Vincent and the Grenadines'
 'Venezuela' 'Virgin Islands' 'Vietnam' 'Vanuatu' 'World' 'Samoa' 'Kosovo'
 'Yemen' 'South Africa' 'Zambia' 'Zimbabwe']


# On recherche le nombre de pays par région
nb_pays_par_region = (
    country.groupby("Region")[["Short Name"]]
    .count()
    .reset_index()
    .sort_values(by="Short Name", ascending=False)
)
nb_pays_par_region


# On visualise le nombre de pays par région
df = (
    country.groupby("Region")[["Short Name"]]
    .count()["Short Name"]
    .sort_values(ascending=True)
)
val = df.tolist()

plt.figure(figsize=(4, 4))
plt.title("Nombre de pays par région", fontname="Arial", fontsize=18, fontweight="bold")
plt.barh(
    y=country.groupby("Region")[["Short Name"]]
    .count()
    .reset_index()
    .sort_values(by="Short Name")["Region"],
    width=country.groupby("Region")[["Short Name"]]
    .count()["Short Name"]
    .sort_values(ascending=True),
)
# plt.grid(axis='x')
# On place les nombre sur le graphe 1 par 1 
plt.text(val[6] - 5.5,  5.8, val[6], color="white", fontweight="bold")
plt.text(val[5] - 5.5,  4.8, val[5], color="white", fontweight="bold")
plt.text(val[4] - 5.5,  3.8, val[4], color="white", fontweight="bold")
plt.text(val[3] - 5.5,  2.8, val[3], color="white", fontweight="bold")
plt.text(val[2] - 5.5,  1.8, val[2], color="white", fontweight="bold")
plt.text(val[1] - 3,  0.8, val[1], color="white", fontweight="bold")
plt.text(val[0] - 2.5, -0.2, val[0], color="white", fontweight="bold")

plt.savefig('assets/graphiques/EDA_1_Nombre de pays par région.jpg', bbox_inches='tight')
plt.show()


# On liste les indicateurs de revenus
country["Income Group"].unique()

array(['High income: nonOECD', 'Low income', 'Upper middle income', nan,
       'Lower middle income', 'High income: OECD'], dtype=object)


# On compte les types de revenu
country["Income Group"].nunique()

5


# On compte le nombre de nan dans la colonne Income Group
print("Il y a",country["Income Group"].isnull().sum(),"données manquantes dans Income Group qui correspondent aux zones ou groupes géographiques")

Il y a 27 données manquantes dans Income Group qui correspondent aux zones ou groupes géographiques


# Afficher les na de la colonne Income Group
country.loc[country["Income Group"].isna() == True, :]


# On calcule le nombre de pays par région
nb_pays_par_groupe_revenu = (
    country.groupby("Income Group")[["Short Name"]]
    .count()
    .reset_index()
    .sort_values(by="Short Name", ascending=False)
)
nb_pays_par_groupe_revenu


# On calcule le nombre de pays par groupe de revenu
nb_pays_par_groupe_revenu = nb_pays_par_groupe_revenu.sort_values(
    by="Short Name", ascending=False
)
nb_pays_par_groupe_revenu


# On visualise le nombre de pays par groupes de revenus
plt.figure(figsize=(15, 7))
plt.barh(
    y=nb_pays_par_groupe_revenu["Income Group"],
    width=country.groupby("Income Group")[["Short Name"]]
    .count()["Short Name"]
    .sort_values(),
)
plt.title(
    "Nombre de pays par groupe de revenu",
    fontname="Arial",
    fontsize=18,
    fontweight="bold",
)
# On place les chiffres sur le graphe avec un boucle
for i in range(5):
    plt.text(
        82 - (nb_pays_par_groupe_revenu.iloc[i, 1]),
        i - 0.1,
        nb_pays_par_groupe_revenu.iloc[4 - i, 1],
        color="white",
        fontweight="bold",
        fontsize=20,
    )

plt.savefig('assets/graphiques/EDA_2_Nombre de pays par groupe de revenu.jpg', bbox_inches='tight')
plt.show()


# On calcule le nombre de pays par groupe de monnaie = monnaie ma plus utilisée
nb_pays_par_groupe_monnaie = (
    country.groupby("Currency Unit")[["Short Name"]]
    .count()
    .reset_index()
    .sort_values(by="Short Name", ascending=False)
)
nb_pays_par_groupe_monnaie.head(6)


# On visualise les 8 monnaies les plus utilisée 
plt.figure(figsize=(8, 4))
plt.barh(
    y=country.groupby("Currency Unit")[["Short Name"]]
    .count()
    .reset_index()
    .sort_values(by="Short Name")["Currency Unit"]
    .tail(8),
    width=country.groupby("Currency Unit")[["Short Name"]]
    .count()["Short Name"]
    .sort_values()
    .tail(8),
)
plt.title(
    "Devises les plus utilisées", fontname="Arial", fontsize=14, fontweight="bold"
)
plt.text(
    21.5,
    6.74,
    nb_pays_par_groupe_monnaie.iloc[0, 1],
    color="white",
    fontweight="bold",
    fontsize=16,
)
plt.text(
    12.5,
    5.74,
    nb_pays_par_groupe_monnaie.iloc[1, 1],
    color="white",
    fontweight="bold",
    fontsize=16,
)
plt.text(
    12.5,
    4.74,
    nb_pays_par_groupe_monnaie.iloc[2, 1],
    color="white",
    fontweight="bold",
    fontsize=16,
)
plt.text(
    5,
    3.74,
    nb_pays_par_groupe_monnaie.iloc[3, 1],
    color="white",
    fontweight="bold",
    fontsize=16,
)
plt.text(
    2.2,
    2.78,
    nb_pays_par_groupe_monnaie.iloc[4, 1],
    color="white",
    fontweight="bold",
    fontsize=16,
)
plt.text(
    2.2,
    1.78,
    nb_pays_par_groupe_monnaie.iloc[5, 1],
    color="white",
    fontweight="bold",
    fontsize=16,
)
plt.text(
    2.2,
    0.78,
    nb_pays_par_groupe_monnaie.iloc[6, 1],
    color="white",
    fontweight="bold",
    fontsize=16,
)
plt.text(
    1.2,
    -0.24,
    nb_pays_par_groupe_monnaie.iloc[7, 1],
    color="white",
    fontweight="bold",
    fontsize=16,
)
plt.savefig('assets/graphiques/EDA_3_Devises les plus utilisées.jpg', bbox_inches='tight')
plt.show()


# Regroupement des pays par groupe de revenus

sns.countplot(
    y="Income Group",
    hue="Region",
    data=country,
    orient="h",
    order=[
        "High income: OECD",
        "High income: nonOECD",
        "Upper middle income",
        "Lower middle income",
        "Low income",
    ],
    palette=sns.color_palette('bright'),
)
plt.xlabel("Nombre de pays", fontsize=12)
plt.ylabel("Groupe de revenus", fontsize=12)
plt.legend(bbox_to_anchor=(1.05, 1), loc=0, borderaxespad=0.0, fontsize=12)
plt.title(
    "Nombre de pays par groupe de revenus par région", fontsize=14, fontweight="bold"
)
plt.grid(False)
plt.savefig('assets/graphiques/EDA_3_Nombre de pays par groupe de revenu et par régions.jpg', bbox_inches='tight')
plt.show()


# Visualisation Regroupements des pays/régions/groupes de revenus pour avoir un visuel unique

fig = plt.figure(figsize=(15, 10))
grid = plt.GridSpec(2, 2, wspace=0.8, hspace=0.3)

# Graphique ligne 1, colonne 1
plt.subplot(grid[0, 0])
# Nombre de pays par région
df = (
    country.groupby("Region")[["Short Name"]]
    .count()["Short Name"]
    .sort_values(ascending=True)
)
val = df.tolist()

plt.title("Nombre de pays par région", fontname="Arial", fontsize=14, fontweight="bold")
plt.barh(
    y=country.groupby("Region")[["Short Name"]]
    .count()
    .reset_index()
    .sort_values(by="Short Name")["Region"],
    width=country.groupby("Region")[["Short Name"]]
    .count()["Short Name"]
    .sort_values(ascending=True),
)
# plt.grid(axis='x')
plt.text(val[6] - 5.5, 5.9, val[6], color="white", fontweight="bold")
plt.text(val[5] - 5.5, 4.9, val[5], color="white", fontweight="bold")
plt.text(val[4] - 5.5, 3.9, val[4], color="white", fontweight="bold")
plt.text(val[3] - 5.5, 2.9, val[3], color="white", fontweight="bold")
plt.text(val[2] - 5.5, 1.9, val[2], color="white", fontweight="bold")
plt.text(val[1] - 3.5, 0.9, val[1], color="white", fontweight="bold")
plt.text(val[0] - 2.5, -0.1, val[0], color="white", fontweight="bold")

# Graphique ligne 1, colonne 2
plt.subplot(grid[0, 1])
# Nombre de pays par groupe de revenus
plt.barh(
    y=nb_pays_par_groupe_revenu["Income Group"],
    width=country.groupby("Income Group")[["Short Name"]]
    .count()["Short Name"]
    .sort_values(),
)
plt.title("Nombre de pays par groupe de revenu", fontweight="bold")
for i in range(5):
    plt.text(
        80 - (nb_pays_par_groupe_revenu.iloc[i, 1]),
        i - 0.1,
        nb_pays_par_groupe_revenu.iloc[4 - i, 1],
        color="white",
        fontweight="bold",
        fontsize=14,
    )

# Graphique ligne 2, sur les 2 colonne
plt.subplot(grid[1, :])
# Regroupement des pays par groupe de revenus

sns.countplot(
    y="Income Group",
    hue="Region",
    data=country,
    orient="h",
    order=[
        "High income: OECD",
        "High income: nonOECD",
        "Upper middle income",
        "Lower middle income",
        "Low income",
    ],
    palette=sns.color_palette("bright"),
)
plt.xlabel("Nombre de pays", fontsize=12)
plt.ylabel("Groupe de revenus", fontsize=12)
plt.legend(bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.0, fontsize=12)
plt.title(
    "Nombre de pays par groupe de revenus par région", fontsize=14, fontweight="bold"
)
plt.grid(False)
plt.savefig('assets/graphiques/EDA_4_Synthèse 3graph.jpg', bbox_inches='tight')
plt.show()


# Visualisation 1 visuel 4 graphiques
plt.figure(figsize=(15, 15), dpi=100)  # taille
sns.set(style="dark")
plt.subplot(4, 2, 1)
plt.title("Distribution - analyse univariée")
df = (
    country.groupby("Region")[["Short Name"]]
    .count()["Short Name"]
    .sort_values(ascending=True)
)
val = df.tolist()
# Graphique 1 : Nombre de pays par Région
plt.title("Nombre de pays par région", fontname="Arial", fontweight="bold", fontsize=14)
plt.barh(
    y=country.groupby("Region")[["Short Name"]]
    .count()
    .reset_index()
    .sort_values(by="Short Name")["Region"],
    width=country.groupby("Region")[["Short Name"]]
    .count()["Short Name"]
    .sort_values(ascending=True),
)
# plt.grid(axis='x')
plt.text(val[6] - 5.5, 5.9, val[6], color="white", fontweight="bold")
plt.text(val[5] - 5.5, 4.9, val[5], color="white", fontweight="bold")
plt.text(val[4] - 5.5, 3.9, val[4], color="white", fontweight="bold")
plt.text(val[3] - 5.5, 2.9, val[3], color="white", fontweight="bold")
plt.text(val[2] - 5.5, 1.9, val[2], color="white", fontweight="bold")
plt.text(val[1] - 3.5, 0.9, val[1], color="white", fontweight="bold")
plt.text(val[0] - 2.5, -0.1, val[0], color="white", fontweight="bold")

# Nombre de pays par groupe de revenu
plt.subplot(4, 2, 2)
plt.barh(
    y=nb_pays_par_groupe_revenu["Income Group"],
    width=country.groupby("Income Group")[["Short Name"]]
    .count()["Short Name"]
    .sort_values(),
)
plt.title(
    "Nombre de pays par groupe de revenu",
    fontname="Arial",
    fontweight="bold",
    fontsize=14,
)
# On place les chiffres sur le graphe avec un boucle
for i in range(5):
    plt.text(
        80 - (nb_pays_par_groupe_revenu.iloc[i, 1]),
        i - 0.1,
        nb_pays_par_groupe_revenu.iloc[4 - i, 1],
        color="white",
        fontweight="bold",
    )

# Monnaies les plus utilisées
plt.subplot(4, 2, 3)
plt.barh(
    y=country.groupby("Currency Unit")[["Short Name"]]
    .count()
    .reset_index()
    .sort_values(by="Short Name")["Currency Unit"]
    .tail(8),
    width=country.groupby("Currency Unit")[["Short Name"]]
    .count()["Short Name"]
    .sort_values()
    .tail(8),
)
plt.title(
    "Monnaies les plus utilisées", fontname="Arial", fontweight="bold", fontsize=14
)
plt.text(
    21.5, 6.74, nb_pays_par_groupe_monnaie.iloc[0, 1], color="white", fontweight="bold"
)
plt.text(
    12.5, 5.74, nb_pays_par_groupe_monnaie.iloc[1, 1], color="white", fontweight="bold"
)
plt.text(
    12.5, 4.74, nb_pays_par_groupe_monnaie.iloc[2, 1], color="white", fontweight="bold"
)
plt.text(
    5, 3.74, nb_pays_par_groupe_monnaie.iloc[3, 1], color="white", fontweight="bold"
)
plt.text(
    2.2, 2.78, nb_pays_par_groupe_monnaie.iloc[4, 1], color="white", fontweight="bold"
)
plt.text(
    2.2, 1.78, nb_pays_par_groupe_monnaie.iloc[5, 1], color="white", fontweight="bold"
)
plt.text(
    2.2, 0.78, nb_pays_par_groupe_monnaie.iloc[6, 1], color="white", fontweight="bold"
)
plt.text(
    1.2, -0.24, nb_pays_par_groupe_monnaie.iloc[7, 1], color="white", fontweight="bold"
)

# Regroupement des pays par groupe de revenus et par région
plt.subplot(4, 2, 4)

sns.countplot(
    y="Income Group",
    hue="Region",
    data=country,
    orient="h",
    order=[
        "High income: OECD",
        "High income: nonOECD",
        "Upper middle income",
        "Lower middle income",
        "Low income",
    ],
    palette=sns.color_palette("bright"),
)
plt.xlabel("Nombre de pays", fontsize=12)
plt.ylabel("Groupe de revenus", fontsize=12)
plt.legend(bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.0, fontsize=12)
plt.title(
    "Nombre de pays par groupe de revenus par région",
    fontname="Arial",
    fontweight="bold",
    fontsize=14,
)


plt.grid(False)

plt.subplots_adjust(left=0.125, bottom=0.1, right=0.9, top=0.9, wspace=1, hspace=0.35)
# bbox_inches='tight'
plt.savefig('assets/graphiques/EDA_5_Synthèse 4graph.jpg', bbox_inches='tight')
plt.show()


# Import données
country_Series = pd.read_csv("assets/data/EdStatsCountry-Series.csv", sep=",")


# On affiche les 5 premières lignes
country_Series.head()


# Premières informations : nom colonnes, nb null, types
country_Series.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 613 entries, 0 to 612
Data columns (total 4 columns):
 #   Column       Non-Null Count  Dtype  
---  ------       --------------  -----  
 0   CountryCode  613 non-null    object 
 1   SeriesCode   613 non-null    object 
 2   DESCRIPTION  613 non-null    object 
 3   Unnamed: 3   0 non-null      float64
dtypes: float64(1), object(3)
memory usage: 19.3+ KB


# Description des variables qualitatives (pas de variable quantitatives dans ce dataframe)
country_Series.describe(exclude=[np.number])


# Utilisation de la fonction def 
info_data(country_Series)

------------------------------------------------------------------------------------------------------
LES DONNES 

Nombre de lignes : 613 lignes
Nombre de colonnes : 4 colonnes
Nombre d'éléments dupliqués dans le dataframe : 0 eléments
Nombre total de données manquantes dans le dataframe : 613 données manquantes sur 2452 (25.0%)


info_data_na(country_Series)

Nombre total de données manquantes dans le dataframe : 613 données manquantes sur 2452 (25.0%)
------------------------------------------------------------------------------------------------------
Visualisation des données manquantes


country_Series["CountryCode"].nunique()

211


# On affiche les pays présents dans CountrySeries mais absent de Country
country[~country['Country Code'].isin(country_Series['CountryCode'].unique())][['Short Name']]


country_Series["SeriesCode"].nunique()

21


country_Series["SeriesCode"].unique()

array(['SP.POP.TOTL', 'SP.POP.GROW', 'NY.GDP.PCAP.PP.CD',
       'NY.GDP.MKTP.PP.KD', 'NY.GNP.MKTP.PP.CD', 'NY.GDP.MKTP.PP.CD',
       'NY.GDP.PCAP.PP.KD', 'NY.GNP.PCAP.PP.CD', 'SP.POP.1564.TO.ZS',
       'SP.POP.TOTL.MA.ZS', 'SP.POP.TOTL.FE.ZS', 'SP.POP.0014.TO.ZS',
       'NY.GNP.PCAP.CD', 'NY.GDP.PCAP.CD', 'NY.GDP.PCAP.KD',
       'SP.POP.1564.MA.IN', 'SP.POP.0014.TO', 'SP.POP.1564.TO',
       'SP.POP.1564.FE.IN', 'SP.POP.0014.MA.IN', 'SP.POP.0014.FE.IN'],
      dtype=object)


country_Series["DESCRIPTION"].unique()

array(['Data sources : United Nations World Population Prospects',
       'Data sources: United Nations World Population Prospects',
       'Estimates are based on regression.',
       'Data sources : Institute of Statistics, Eurostat',
       'Data sources : Australian Bureau of Statistics',
       'Including Other Territories comprising Jervis Bay Territory, Christmas Island and the Cocos (Keeling) Islands. ',
       'Data sources: Australian Bureau of Statistics',
       'Data sources : Eurostat',
       'Data sources : Eurostat, State Statistical Committee, United Nations World Population Prospects',
       'Data sources: Eurostat, State Statistical Committee, United Nations World Population Prospects',
       'Data sources : National Statistical Committee of the Republic of Belarus',
       'Data sources: National Statistical Committee of the Republic of Belarus',
       'Data sources : Bermuda Government Department of Statistics',
       'Data sources: Bermuda Government Department of Statistics',
       'Data sources : Statistics Canada',
       'Data sources: Statistics Canada',
       'Refers to Guernsey, and Jersey.',
       'Data sources : National Bureau of Statistics',
       'Excluding Hong Kong SAR, Macao SAR, Taiwan.',
       'Data sources: National Bureau of Statistics',
       'Data are for areas under the effective control of the Government of the Republic of Cyprus.',
       'Including Northern-Cyprus.',
       'Data sources: Eurostat, United Nations World Population Prospects. Including the French overseas departments of French Guiana, Guadeloupe, Martinique, Mayotte, and Réunion.',
       'Including the four French overseas departments (French Guyana, Guadeloupe, Martinique and Réunion)',
       'Data sources : Eurostat, United Nations World Population Prospects. Including the French overseas departments of French Guiana, Guadeloupe, Martinique, Mayotte, and Réunion.',
       'Data sources: Eurostat',
       'Data sources : GeoStats. GeoStats. Adjusted to mid-year population using geometric mean for the years where mid-year population figures are not available. Excluding Abkhazia and South Ossetia.',
       'Including Abkhazia and South Ossetia.',
       'Excludes Abkhazia and South Ossetia.',
       'Derived using ratio of age group from WPP and World Bank population.',
       'Data sources: GeoStats. GeoStats. Adjusted to mid-year population using geometric mean for the years where mid-year population figures are not available. Excluding Abkhazia and South Ossetia.',
       'Excluding Abkhazia and South Ossetia.',
       'Data sources: Statistics Greenland',
       'Data sources : Statistics Greenland',
       'Data sources: Census and Statistics Department',
       'Data sources : Census and Statistics Department',
       'Data sources: Croatian Bureau of Statistics, Council of Europe',
       'Data sources : Croatian Bureau of Statistics, Council of Europe',
       'Data sources: Central Bureau of Statistics',
       'Data sources : Central Bureau of Statistics',
       'Data sources: Ministry of Internal Affairs and Communications, Statistics Bureau; United Nations Population and Vital Statistics Report',
       'Data sources : Ministry of Internal Affairs and Communications, Statistics Bureau; United Nations Population and Vital Statistics Report',
       'Data sources : Statistical Agency of Kazakhstan, 1979 Census, 1989 Census, United Nations World Population Prospects',
       'Data sources: Statistical Agency of Kazakhstan, 1979 Census, 1989 Census, United Nations World Population Prospects',
       'Data sources : National Statistical Committee, United Nations World Population Prospects',
       'Data sources: National Statistical Committee, United Nations World Population Prospects',
       'Data sources : Korean Statistical Information Service',
       'Data sources: Korean Statistical Information Service',
       "Data sources : Registrar General's Department, Department of Census and Statistics",
       "Data sources: Registrar General's Department, Department of Census and Statistics",
       'Includes Former Spanish Sahara.',
       'Morocco only. (Excluding former Spanish Sahara)',
       "Data sources : National Bureau of Statistics. Excluding Transnistria. For 1950-94, World Bank estimates using UN World Population Prospects' growth rates of whole Moldova.",
       'Excludes Transnistria.', 'Including Transnistria.',
       "Data sources: National Bureau of Statistics. Excluding Transnistria. For 1950-94, World Bank estimates using UN World Population Prospects 2010 Revision's growth rates of whole Moldova.",
       'Excluding Transnistria.',
       'Data sources : For 1960-1989, United Nations World Population Prospects. For 1990 onwards, 1990, 1995, 2000, 2006, 2014 censuses, and data on expatriate employment from Maldives National Bureau of Statistics. The census data are adjusted for mid-year and for foreign population.',
       'Data sources: For 1960-1989, United Nations World Population Prospects. For 1990 onwards, 1990, 1995, 2000, 2006, 2014 censuses, and data on expatriate employment from Maldives National Bureau of Statistics. The census data are adjusted for mid-year and for foreign population.',
       'Data sources : For years after 2000, Statistical Office of Montenegro. For years 1960-1999, growth rates of United Nations World Population Prospects were used to estimate backwards.',
       'Data sources : Statistics Mauritius, United Nations World Population Prospects',
       'Including Agalega and St. Brandon.',
       'Data sources: Central Statistics Office, United Nations Population and Vital Statistics Report, United Nations World Population Prospects',
       'Excluding Agalega and St. Brandon.',
       'Including Sabah and Sarawak.',
       'Data sources : Institute of Statistics and Economic Studies',
       'Data sources: Institute of Statistics and Economic Studies, 1983 Census, 1989 Census, 1996 Census',
       'Data sources: Statistics New Zealand',
       'Data sources : Statistics New Zealand',
       'Data sources : U.S. Census Bureau, 1990 Census',
       'Data sources: U.S. Census Bureau, 1990 Census',
       'Data sources : World Bank estimates based on data from Palestinian Central Bureau of Statistics, excluding E. Jerusalem.',
       'Excluding East Jerusalem.',
       'Data sources: World Bank estimates based on data from Palestinian Central Bureau of Statistics, excluding E. Jerusalem.',
       'Data sources : Russian Federation Federal State Statistics Service (its January 1st populations were adjusted to mid-year populations by using geometric mean), 1979 Census, 1989 Census.',
       'Data sources : United Nations World Population Prospects. Estimates are for Sudan excluding South Sudan.',
       'Data sources: Department of Statistics Singapore',
       'Data sources : Department of Statistics Singapore',
       'Data sources: Statistical Office of the Republic of Serbia, Eurostat, TransMONEE, Council of Europe. Excluding Kosovo for all years 1950-2011',
       'Including Kosovo.',
       'Data sources : Statistical Office of the Republic of Serbia, Eurostat, TransMONEE, Council of Europe. Note : Excluding Kosovo.',
       'Excluding Kosovo.', 'Covers mainland Tanzania only.',
       'Including Zanzibar.',
       'Data sources : State Statistics Committee of Ukraine, United Nations World Population Prospects, Council of Europe, United Nations Population and Vital Statistics Report',
       'Data sources: State Statistics Committee of Ukraine, United Nations World Population Prospects, Council of Europe, United Nations Population and Vital Statistics Report',
       'Data sources : U.S. Census Bureau',
       'Data sources: U.S. Census Bureau',
       'Data sources : State Committee of the Republic of Uzbekistan on Statistics, United Nations World Population Prospects',
       'Based on official exchange rates.',
       'Data sources : U.S. Census Bureau, United Nations Population and Vital Statistics Report',
       'Data sources: U.S. Census Bureau, United Nations Population and Vital Statistics Report',
       'Data sources: General Statistics Office of Vietnam',
       'Data sources : General Statistics Office of Vietnam',
       "Data sources : World Bank's Kosovo coutnry office, Kosovo Agency of Statistics, U.S. Census Bureau",
       "Data sources: World Bank's Kosovo coutnry office, Kosovo Agency of Statistics, U.S. Census Bureau",
       'Data sources : Statistics South Africa, United Nations World Population Prospects'],
      dtype=object)


Series_code = country_Series.groupby('SeriesCode')[['CountryCode']].count().reset_index().sort_values(by='CountryCode')['SeriesCode']
nb_pays_serie_code=country_Series.groupby('SeriesCode')[['CountryCode']].count()['CountryCode'].sort_values(ascending=True)

# Graphique : Nombre de pays par indicateurs
plt.barh(Series_code,nb_pays_serie_code)
plt.title("Repartition du Nombre de pays par indicateur",weight='bold')
for i, var in enumerate(nb_pays_serie_code):
    plt.text(var+1,i-0.5,str(var),color='steelblue',fontweight='bold')
plt.savefig('assets/graphiques/EDA_6_Répartition pays par indicateurs.jpg', bbox_inches='tight')


# Graphique : Répartition des 10 sources de données les plus utilisées
description = country_Series.groupby('DESCRIPTION')[['CountryCode']].count().reset_index().sort_values(by='CountryCode')['DESCRIPTION'].tail(10)
nb_pays_description = country_Series.groupby('DESCRIPTION')[['CountryCode']].count()['CountryCode'].sort_values().tail(10)
                
plt.barh(description,nb_pays_description)
plt.title("Repartition du Nombre de données par nombre de pays par sources",weight='bold')
for i, var in enumerate(nb_pays_description):
    plt.text(var+1,i-0.2,str(var),color='steelblue',fontweight='bold')
plt.savefig('assets/graphiques/EDA_7_Répartition source pays.jpg', bbox_inches='tight')


# Création d'un visuel unique concernant les indicateurs et les descriptions
fig, ax = plt.subplots(1, 2, figsize=(20, 7)) 

# Regroupement des pays par code indicateur

ax1=ax[0]
ax1.barh(y = Series_code, width=nb_pays_serie_code, color='SteelBlue')
for i, v in enumerate(nb_pays_serie_code):
    ax1.text(v + 3, i - .22, str(v), color='SteelBlue', fontsize=10, fontweight='bold')
ax1.set_title('Nombre de pays par code indicateur', fontsize=12,fontweight='bold')
ax1.set_xlabel('Nombre de pays', fontsize=10)
ax1.set_ylabel('Codes indicateurs', fontsize=10)
ax1.tick_params(labelsize=10)
ax1.set_xlim([0, 250])
# ax1.grid(True)

# Graphique 2 : Top 10 descriptions (sources)
ax2=ax[1]
ax2.barh(y = description, width=nb_pays_description, color='SteelBlue')
for i, v in enumerate(nb_pays_description):
    ax2.text(v + 3, i + .01, str(v), color='SteelBlue', fontsize=10, fontweight='bold')
ax2.set_title('Nombre de pays pour les tops 10 descriptions', fontsize=12,fontweight='bold')
ax2.set_xlabel('Nombre de pays', fontsize=10)
ax2.set_ylabel('Sources des codes indicateurs', fontsize=10)
ax2.tick_params(labelsize=10)
ax2.set_xlim([0, 180])
# ax2.grid(True)
plt.savefig('assets/graphiques/EDA_8_Synthèse 2graph répartition indicatreur et sources.jpg', bbox_inches='tight')
fig.tight_layout()
plt.show()


# Import données
footNote = pd.read_csv("assets/data/EdStatsFootNote.csv", sep=",")


footNote.head()


footNote.columns

Index(['CountryCode', 'SeriesCode', 'Year', 'DESCRIPTION', 'Unnamed: 4'], dtype='object')


info_data(footNote)

------------------------------------------------------------------------------------------------------
LES DONNES 

Nombre de lignes : 643638 lignes
Nombre de colonnes : 5 colonnes
Nombre d'éléments dupliqués dans le dataframe : 0 eléments
Nombre total de données manquantes dans le dataframe : 643638 données manquantes sur 3218190 (20.0%)


footNote.dtypes

CountryCode     object
SeriesCode      object
Year            object
DESCRIPTION     object
Unnamed: 4     float64
dtype: object


# Résumé des variables qualitatives
country.describe(exclude=[np.number])


info_data_na(footNote)

Nombre total de données manquantes dans le dataframe : 643638 données manquantes sur 3218190 (20.0%)
------------------------------------------------------------------------------------------------------
Visualisation des données manquantes


# Liste des noms des pays présents dans Country et qui ne sont pas présents dans le jeu de données EdStatsCountry-Series.csv
country[~country["Country Code"].isin(country_Series["CountryCode"])][
    ["Short Name"]
].sort_values(by="Short Name")


list_description = footNote['DESCRIPTION'].unique().tolist()


# On affiche les années
footNote['Year'].unique()

array(['YR2001', 'YR2005', 'YR2000', 'YR2004', 'YR2006', 'YR2003',
       'YR1999', 'YR2008', 'YR2002', 'YR2007', 'YR2017', 'YR2010',
       'YR2009', 'YR2011', 'YR2012', 'YR2015', 'YR2014', 'YR1970',
       'YR1971', 'YR1972', 'YR1973', 'YR1974', 'YR1975', 'YR1976',
       'YR1977', 'YR1978', 'YR1979', 'YR1980', 'YR1981', 'YR1982',
       'YR1983', 'YR1984', 'YR1985', 'YR1986', 'YR1987', 'YR1988',
       'YR1989', 'YR1990', 'YR1991', 'YR1992', 'YR1993', 'YR1994',
       'YR1995', 'YR1996', 'YR1997', 'YR1998', 'YR2013', 'YR2016',
       'YR2020', 'YR2025', 'YR2030', 'YR2035', 'YR2040', 'YR2045',
       'YR2050', 'yr2012'], dtype=object)


# On supprime les 2 premières lettres pour avoir uniquement l'année
footNote['Year']=footNote['Year'].apply(lambda x : x[2:])
footNote['Year'].unique()

array(['2001', '2005', '2000', '2004', '2006', '2003', '1999', '2008',
       '2002', '2007', '2017', '2010', '2009', '2011', '2012', '2015',
       '2014', '1970', '1971', '1972', '1973', '1974', '1975', '1976',
       '1977', '1978', '1979', '1980', '1981', '1982', '1983', '1984',
       '1985', '1986', '1987', '1988', '1989', '1990', '1991', '1992',
       '1993', '1994', '1995', '1996', '1997', '1998', '2013', '2016',
       '2020', '2025', '2030', '2035', '2040', '2045', '2050'],
      dtype=object)


# On affiche les données en fonction des années
footNote.groupby('Year').count()


footNote['Year'].sort_values().unique()

array(['1970', '1971', '1972', '1973', '1974', '1975', '1976', '1977',
       '1978', '1979', '1980', '1981', '1982', '1983', '1984', '1985',
       '1986', '1987', '1988', '1989', '1990', '1991', '1992', '1993',
       '1994', '1995', '1996', '1997', '1998', '1999', '2000', '2001',
       '2002', '2003', '2004', '2005', '2006', '2007', '2008', '2009',
       '2010', '2011', '2012', '2013', '2014', '2015', '2016', '2017',
       '2020', '2025', '2030', '2035', '2040', '2045', '2050'],
      dtype=object)


# Visualiser la répartion du nombre d'indicateurs par pays et année
plt.figure(figsize=(15,5))
plt.title("Nombres d'indicateurs par pays et années",fontweight='bold')
sns.barplot(data=footNote.groupby('Year').count().reset_index(),y='CountryCode',x='Year')  
plt.xticks(rotation=90)
plt.savefig("assets/graphiques/EDA_9_Nombres d'indicateurs par pays et années.jpg", bbox_inches='tight')
plt.show()


# Import données
serie = pd.read_csv("assets/data/EdStatsSeries.csv", sep=",")


serie.head()


info_data(serie)

------------------------------------------------------------------------------------------------------
LES DONNES 

Nombre de lignes : 3665 lignes
Nombre de colonnes : 21 colonnes
Nombre d'éléments dupliqués dans le dataframe : 0 eléments
Nombre total de données manquantes dans le dataframe : 55203 données manquantes sur 76965 (71.72%)


serie.dtypes

Series Code                             object
Topic                                   object
Indicator Name                          object
Short definition                        object
Long definition                         object
Unit of measure                        float64
Periodicity                             object
Base Period                             object
Other notes                             object
Aggregation method                      object
Limitations and exceptions              object
Notes from original source             float64
General comments                        object
Source                                  object
Statistical concept and methodology     object
Development relevance                   object
Related source links                    object
Other web links                        float64
Related indicators                     float64
License Type                           float64
Unnamed: 20                            float64
dtype: object


info_data_na(serie)

Nombre total de données manquantes dans le dataframe : 55203 données manquantes sur 76965 (71.72%)
------------------------------------------------------------------------------------------------------
Visualisation des données manquantes


# On affiche le pourcentage de données manquantes par colonne classée dans l'ordre croissant
na_serie = serie.isnull().mean()*100
na_serie.sort_values()

Series Code                              0.000000
Topic                                    0.000000
Indicator Name                           0.000000
Long definition                          0.000000
Source                                   0.000000
Short definition                        41.173261
Other notes                             84.938608
Base Period                             91.432469
Related source links                    94.133697
Periodicity                             97.298772
Aggregation method                      98.717599
Statistical concept and methodology     99.372442
Limitations and exceptions              99.618008
General comments                        99.618008
Development relevance                   99.918145
Notes from original source             100.000000
License Type                           100.000000
Unit of measure                        100.000000
Other web links                        100.000000
Related indicators                     100.000000
Unnamed: 20                            100.000000
dtype: float64


serie['Series Code'].head()

0    BAR.NOED.1519.FE.ZS
1       BAR.NOED.1519.ZS
2    BAR.NOED.15UP.FE.ZS
3       BAR.NOED.15UP.ZS
4    BAR.NOED.2024.FE.ZS
Name: Series Code, dtype: object


serie['Series Code'].nunique()

3665


# Afficher 5 dernieres valeurs 
serie['Topic'].tail()

3660    Expenditures
3661    Expenditures
3662    Expenditures
3663    Expenditures
3664    Expenditures
Name: Topic, dtype: object


# Lister les items
list_item = serie['Topic'].unique()
# Visualiser les 20 premiers items
list_item[:20]

array(['Attainment', 'Education Equality',
       'Infrastructure: Communications', 'Learning Outcomes',
       'Economic Policy & Debt: National accounts: US$ at current prices: Aggregate indicators',
       'Economic Policy & Debt: National accounts: US$ at constant 2010 prices: Aggregate indicators',
       'Economic Policy & Debt: Purchasing power parity',
       'Economic Policy & Debt: National accounts: Atlas GNI & GNI per capita',
       'Teachers', 'Education Management Information Systems (SABER)',
       'Early Child Development (SABER)',
       'Engaging the Private Sector (SABER)',
       'School Health and School Feeding (SABER)',
       'School Autonomy and Accountability (SABER)',
       'School Finance (SABER)', 'Student Assessment (SABER)',
       'Teachers (SABER)', 'Tertiary Education (SABER)',
       'Workforce Development (SABER)', 'Literacy'], dtype=object)


serie['Topic'].nunique()

37


serie['Indicator Name'].nunique()

3665


list_indicator = serie['Indicator Name'].unique().tolist()
list_indicator[350:400]

['Barro-Lee: Average years of tertiary schooling, age 55-59, total',
 'Barro-Lee: Average years of tertiary schooling, age 55-59, female',
 'Barro-Lee: Average years of tertiary schooling, age 60-64, total',
 'Barro-Lee: Average years of tertiary schooling, age 60-64, female',
 'Barro-Lee: Average years of tertiary schooling, age 65-69, total',
 'Barro-Lee: Average years of tertiary schooling, age 65-69, female',
 'Barro-Lee: Average years of tertiary schooling, age 70-74, total',
 'Barro-Lee: Average years of tertiary schooling, age 70-74, female',
 'Barro-Lee: Average years of tertiary schooling, age 75+, total',
 'Barro-Lee: Average years of tertiary schooling, age 75+, female',
 'DHS: Gross attendance rate. Post Secondary',
 'DHS: Gross attendance rate. Post Secondary. Female',
 'DHS: Gross attendance rate. Post Secondary. Male',
 'DHS: Gross attendance rate. Post Secondary. Quintile 1',
 'DHS: Gross attendance rate. Post Secondary. Quintile 2',
 'DHS: Gross attendance rate. Post Secondary. Quintile 3',
 'DHS: Gross attendance rate. Post Secondary. Quintile 4',
 'DHS: Gross attendance rate. Post Secondary. Quintile 5',
 'DHS: Gross attendance rate. Post Secondary. Rural',
 'DHS: Gross attendance rate. Post Secondary. Urban',
 'DHS: Net attendance rate. Primary',
 'DHS: Net attendance rate. Primary. Female',
 'DHS: Net attendance rate. Primary. Male',
 'DHS: Net attendance rate. Primary. Quintile 1',
 'DHS: Net attendance rate. Primary. Quintile 2',
 'DHS: Net attendance rate. Primary. Quintile 3',
 'DHS: Net attendance rate. Primary. Quintile 4',
 'DHS: Net attendance rate. Primary. Quintile 5',
 'DHS: Net attendance rate. Primary. Rural',
 'DHS: Net attendance rate. Primary. Urban',
 'DHS: Net attendance rate. Secondary',
 'DHS: Net attendance rate. Secondary. Female',
 'DHS: Net attendance rate. Secondary. Male',
 'DHS: Net attendance rate. Secondary. Quintile 1',
 'DHS: Net attendance rate. Secondary. Quintile 2',
 'DHS: Net attendance rate. Secondary. Quintile 3',
 'DHS: Net attendance rate. Secondary. Quintile 4',
 'DHS: Net attendance rate. Secondary. Quintile 5',
 'DHS: Net attendance rate. Secondary. Rural',
 'DHS: Net attendance rate. Secondary. Urban',
 'DHS: Net intake rate for the first grade of primary education',
 'DHS: Net intake rate for the first grade of primary education. Female',
 'DHS: Net intake rate for the first grade of primary education. Male',
 'DHS: Net intake rate for the first grade of primary education. Quintile 1',
 'DHS: Net intake rate for the first grade of primary education. Quintile 2',
 'DHS: Net intake rate for the first grade of primary education. Quintile 3',
 'DHS: Net intake rate for the first grade of primary education. Quintile 4',
 'DHS: Net intake rate for the first grade of primary education. Quintile 5',
 'DHS: Net intake rate for the first grade of primary education. Rural',
 'DHS: Net intake rate for the first grade of primary education. Urban']


# Nombre de variable dans serie
serie['Long definition'].count()

3665


# On affiche les définitions longues
serie['Long definition'].head(2)

0    Percentage of female population age 15-19 with...
1    Percentage of population age 15-19 with no edu...
Name: Long definition, dtype: object


# On affiche les définitions longues 
serie['Short definition'].count()

2156


# Nombres d'indicateurs pour les 10 items les plus renseignés
items = serie.groupby('Topic')['Series Code'].count().reset_index().sort_values(by="Series Code",ascending=False)
items


# Import données
data = pd.read_csv("assets/data/EdStatsData.csv", sep=",")


data.head()


# Nombre de type de variable
values = data.dtypes.value_counts()
nb_tt = values.sum()
percent = round(values / nb_tt*100,2)
# print(values)
# print(nb_tt)
tab_type = pd.concat([values, percent], axis=1)
tab_type.columns = ['Nombre par type de variable'
                 , '% des types de variable']  # création des colonnes
display(tab_type[tab_type['Nombre par type de variable'] != 0]
        .sort_values('% des types de variable', ascending = False)# trier les valeurs par ordre decroissant
        .style.background_gradient("Blues")) # choix couleurs radial


plt.pie(x = data.dtypes.value_counts(),
        labels = data.dtypes.value_counts().index,
       )
plt.title('Répartion des types de variables',fontweight='bold')
plt.show()


# Résumé des variables qualitatives
data.describe(exclude=[np.number])


# Résumé des variables quantitatives
data.describe()


info_data(data)

------------------------------------------------------------------------------------------------------
LES DONNES 

Nombre de lignes : 886930 lignes
Nombre de colonnes : 70 colonnes
Nombre d'éléments dupliqués dans le dataframe : 0 eléments
Nombre total de données manquantes dans le dataframe : 53455179 données manquantes sur 62085100 (86.1%)


info_data_na(data)

Nombre total de données manquantes dans le dataframe : 53455179 données manquantes sur 62085100 (86.1%)
------------------------------------------------------------------------------------------------------
Visualisation des données manquantes


# data['Indicator Name'].unique().tolist()
# Nombre d'indicateurs Indicator Name
data['Indicator Name'].nunique()

3665


data['Country Code'].nunique()

242


data['Country Name'].nunique()

242


# 3665 indicateurs * 242 pays et zones = notre nb le ligne
data['Indicator Code'].nunique() * data['Country Name'].nunique() == data.shape[0]

True


# On a le même nombre d'indicateurs dans data et dans serie
serie['Series Code'].nunique() == data['Indicator Code'].nunique()

True


data['Country Code'].nunique() == country['Country Code'].nunique()
# Les code ISO-3 présents ne sont pas les mêmes dans ces 2 fichiers

False


# On recheche la liste des pays qui sont différents entre ces 2 fichiers
df_pays_en_trop = data[~data['Country Code'].isin(country['Country Code'])][['Country Name','Country Code']]
df_pays_en_trop.nunique()

Country Name    1
Country Code    1
dtype: int64


# Pays en trop dans EdStatsData.csv est 'British Virgin Islands'
df_pays_en_trop.drop_duplicates(inplace=True)
df_pays_en_trop


data['Country Code'].nunique() == country_Series['CountryCode'].nunique()
# Les code ISO-3 présents ne sont pas les mêmes dans ces 2 fichiers

False


# On recheche la liste des pays qui sont différents entre ces 2 fichiers
df_pays_en_trop = data[~data['Country Code'].isin(country_Series['CountryCode'])][['Country Name','Country Code']]
# nombre de Country Code en trop
df_pays_en_trop.nunique()

Country Name    31
Country Code    31
dtype: int64


df_pays_en_trop.drop_duplicates(inplace=True)
df_pays_en_trop


# Zones et indication revennus : suppression des subdivision de pays
liste_zones_et_income=df_pays_en_trop['Country Code'].to_list()
print("liste des codes ISO-3 des zones pays et incomes :",liste_zones_et_income)

liste des codes ISO-3 des zones pays et incomes : ['ARB', 'EAS', 'EAP', 'EMU', 'ECS', 'ECA', 'EUU', 'HPC', 'HIC', 'LCN', 'LAC', 'LDC', 'LMY', 'LIC', 'LMC', 'MEA', 'MNA', 'MIC', 'NAC', 'OED', 'SAS', 'SSF', 'SSA', 'UMC', 'WLD', 'VGB', 'CUW', 'NRU', 'SXM', 'SSD', 'MAF']


# Répartition du nombre de données renseignées par années pour tous les indicateurs et tous les pays
plt.figure(figsize=(15, 10))
donnees_dispo = data.loc[:,'1970':'2100'].notnull().sum()
ax1 = donnees_dispo.plot(kind='bar', color='SteelBlue')
ax1.set_title('Répartition du nombre de données par année',fontweight='bold')
ax1.set_ylabel('Nombre de données')
plt.savefig("assets/graphiques/EDA_10_Répartition du nombre de données par année.jpg", bbox_inches='tight')
ax1.grid(False)


# On créer des colonnes par decennies
data['1970s'] = data.loc[:,'1970':'1979'].mean(1)
data['1980s'] = data.loc[:,'1980':'1989'].mean(1)
data['1990s'] = data.loc[:,'1990':'1999'].mean(1)
data['2000s'] = data.loc[:,'2000':'2009'].mean(1)
data['2010s'] = data.loc[:,'2010':'2017'].mean(1)
data['2020s'] = data.loc[:,['2020','2025']].mean(1)
data['2030s'] = data.loc[:,['2030','2035']].mean(1)
data['2040s'] = data.loc[:,['2040','2045']].mean(1)
data['2050s'] = data.loc[:,['2050','2055']].mean(1)
data['2060s'] = data.loc[:,['2060','2065']].mean(1)
data['2070s'] = data.loc[:,['2070','2075']].mean(1)
data['2080s'] = data.loc[:,['2080','2085']].mean(1)
data['2090s'] = data.loc[:,['2090','2095']].mean(1)
data['2100s'] = data.loc[:,'2050']


# On crée df_decennies de data pour travailler dessus et regrouper les années par decennies
#ddata_decennie pour annalyser les décennies
df_decennies = data.copy()


# On affiche les colonnes
df_decennies.columns
# Il manque les années 2018,2019
# A partir de 2020, 2 colonnes par décennie

Index(['Country Name', 'Country Code', 'Indicator Name', 'Indicator Code',
       '1970', '1971', '1972', '1973', '1974', '1975', '1976', '1977', '1978',
       '1979', '1980', '1981', '1982', '1983', '1984', '1985', '1986', '1987',
       '1988', '1989', '1990', '1991', '1992', '1993', '1994', '1995', '1996',
       '1997', '1998', '1999', '2000', '2001', '2002', '2003', '2004', '2005',
       '2006', '2007', '2008', '2009', '2010', '2011', '2012', '2013', '2014',
       '2015', '2016', '2017', '2020', '2025', '2030', '2035', '2040', '2045',
       '2050', '2055', '2060', '2065', '2070', '2075', '2080', '2085', '2090',
       '2095', '2100', 'Unnamed: 69', '1970s', '1980s', '1990s', '2000s',
       '2010s', '2020s', '2030s', '2040s', '2050s', '2060s', '2070s', '2080s',
       '2090s', '2100s'],
      dtype='object')


df_decennies.drop(['1970', '1971', '1972', '1973', '1974', '1975', '1976', '1977', '1978',
       '1979', '1980', '1981', '1982', '1983', '1984', '1985', '1986', '1987',
       '1988', '1989', '1990', '1991', '1992', '1993', '1994', '1995', '1996',
       '1997', '1998', '1999', '2000', '2001', '2002', '2003', '2004', '2005',
       '2006', '2007', '2008', '2009', '2010', '2011', '2012', '2013', '2014',
       '2015', '2016', '2017', '2020', '2025', '2030', '2035', '2040', '2045',
       '2050', '2055', '2060', '2065', '2070', '2075', '2080', '2085', '2090',
       '2095', '2100'], axis=1, inplace=True)


# Nombre total de données - le nombre de données manquante par décénies
decennies = ['1970s', '1980s', '1990s', '2000s', '2010s', '2020s', '2030s', '2040s','2050s', '2060s', '2070s', '2080s', '2090s', '2100s']
plt.figure(figsize=(16, 4))
plt.title('Nombre de données significatives par décennie', size=15,fontweight='bold')
sns.set_style('whitegrid')
sns.barplot(y = df_decennies.shape[0]-df_decennies[decennies].isna().sum().values, x = df_decennies[decennies].isna().sum().index)
plt.savefig("assets/graphiques/EDA_11_Répartition du nombre de données.jpg", bbox_inches='tight')


plt.figure(figsize=(8, 6))
donnees = df_decennies.loc[:,'1970s':'2010s'].notnull().sum()
ax1 = donnees.plot(kind='bar', color='SteelBlue')
ax1.set_title('Répartition du nombre de données par décennies',weight='bold')
ax1.set_ylabel('Nombre de données')
plt.xticks(rotation = 0)
for i in range(5):
    plt.text(i-0.23,donnees[i]+2000, round(donnees[i],1)) # on place en fonction des coordonnées
ax1.grid(False)
plt.savefig("assets/graphiques/EDA_12_Répartition du nombre de données par decennies.jpg", bbox_inches='tight')
plt.show()


data.columns

Index(['Country Name', 'Country Code', 'Indicator Name', 'Indicator Code',
       '1970', '1971', '1972', '1973', '1974', '1975', '1976', '1977', '1978',
       '1979', '1980', '1981', '1982', '1983', '1984', '1985', '1986', '1987',
       '1988', '1989', '1990', '1991', '1992', '1993', '1994', '1995', '1996',
       '1997', '1998', '1999', '2000', '2001', '2002', '2003', '2004', '2005',
       '2006', '2007', '2008', '2009', '2010', '2011', '2012', '2013', '2014',
       '2015', '2016', '2017', '2020', '2025', '2030', '2035', '2040', '2045',
       '2050', '2055', '2060', '2065', '2070', '2075', '2080', '2085', '2090',
       '2095', '2100', 'Unnamed: 69', '1970s', '1980s', '1990s', '2000s',
       '2010s', '2020s', '2030s', '2040s', '2050s', '2060s', '2070s', '2080s',
       '2090s', '2100s'],
      dtype='object')


# On fait une jointure de country et data
data = data.merge(right=country[['Country Code','Region','Income Group']], 
                  on='Country Code',
                  how='left')


# data.drop(['Unnamed: 69'], axis=1, inplace=True)


data.columns

Index(['Country Name', 'Country Code', 'Indicator Name', 'Indicator Code',
       '1970', '1971', '1972', '1973', '1974', '1975', '1976', '1977', '1978',
       '1979', '1980', '1981', '1982', '1983', '1984', '1985', '1986', '1987',
       '1988', '1989', '1990', '1991', '1992', '1993', '1994', '1995', '1996',
       '1997', '1998', '1999', '2000', '2001', '2002', '2003', '2004', '2005',
       '2006', '2007', '2008', '2009', '2010', '2011', '2012', '2013', '2014',
       '2015', '2016', '2017', '2020', '2025', '2030', '2035', '2040', '2045',
       '2050', '2055', '2060', '2065', '2070', '2075', '2080', '2085', '2090',
       '2095', '2100', 'Unnamed: 69', '1970s', '1980s', '1990s', '2000s',
       '2010s', '2020s', '2030s', '2040s', '2050s', '2060s', '2070s', '2080s',
       '2090s', '2100s', 'Region', 'Income Group'],
      dtype='object')


# Repartition des données par région pour la décénnie 2010s
target_decade = ['2010','2011','2012','2013','2014','2015','2016','2017']
data.groupby('Region')[target_decade].count().sum(axis=1).sort_values().reset_index()


# Taux de remplissage
nb_tt = data.shape[0]
nb_nonnull=data.copy()[target_decade].count()
# Taux de données non nulle par année pour 2010s
df_2010s = pd.DataFrame({'annees':nb_nonnull.index, 'nb_nonnull':nb_nonnull.values})
# % donnee null par 
df_2010s['%_nonnull']=round((df_2010s['nb_nonnull'])*100/nb_tt,2)
# Nb nan pour les années cibles
df_2010s['%_NaN']=round(100-df_2010s['%_nonnull'],2)


# Definir les années comme index 
df_2010s = df_2010s.set_index('annees')


df_2010s


# Graphique Taux de remplissage
from matplotlib import rc
plt.figure(figsize=(8, 6))
rc('font', weight='bold')
barWidth = 0.9
annees = ['2010','2011','2012','2013','2014','2015','2016','2017']
taux_remplissage=df_2010s['%_nonnull']
taux_NaN=df_2010s['%_NaN']
plt.bar(annees, taux_remplissage, barWidth, label='Données', color='steelblue')
plt.bar(annees, taux_NaN, barWidth, bottom=taux_remplissage, label='NaN', color='gray')

plt.ylabel('%')
names = ['2010','2011','2012','2013','2014','2015','2016','2017']
r = [0,1,2,3,4,5,6,7]
plt.xticks(r, names, fontweight='bold')
plt.title('Taux de remplissage entre 2010 et 2017',fontweight='bold',size=15)
plt.legend(bbox_to_anchor=(1, 1),loc = 2)

for i, var in enumerate(df_2010s['%_nonnull']):
    plt.text(i-0.4,8,str(var)+"%",color='white',fontweight='bold')
plt.grid(False)
plt.savefig("assets/graphiques/EDA_13_Taux de remplissage 2010-2017.jpg", bbox_inches='tight')
plt.show()


df_taux_indicateurs = data.copy()
annees =['2010','2011','2012','2013','2014','2015']
df_taux_indicateurs=df_taux_indicateurs[~df_taux_indicateurs['Country Code'].isin(liste_zones_et_income)]


# Compter le nombre d'indicateur présents par années
df_taux_indicateurs = df_taux_indicateurs.groupby('Indicator Code')[annees].count()


df_taux_indicateurs['nb_indicateur']=df_taux_indicateurs[annees].sum(axis=1)


df_taux_indicateurs['%'] = (df_taux_indicateurs['nb_indicateur']/(data['Country Name'].nunique()*len(annees)))*100
df_taux_indicateurs.head()


# On crée une colonne total = nombre d'indicateur pour tous les pays par indicateur
# df_taux_indicateurs.sort_values(by ='%',ascending=False)


print(f"Nombre de Country Name : {data['Country Name'].nunique()}")
print(f"Nombre d'années : {len(annees)}")

Nombre de Country Name : 242
Nombre d'années : 6


df_taux_indicateurs[['nb_indicateur','%']].describe().T


# Taux de remplissage
sns.kdeplot(data=df_taux_indicateurs, x="%")
plt.grid(False)
plt.xlim([0,100])
plt.title('Taux de remplissage des indicateurs',weight='bold',size=15)
plt.savefig("assets/graphiques/PREPROCESSING_1_Taux de remplissage des indicateurs.jpg", bbox_inches='tight')
plt.show()


# On garde tous les indicateur >50 % ==> Seuil = 50%
df_indicateurs_reduit = df_taux_indicateurs[df_taux_indicateurs['%']>80]
df_indicateurs_reduit['Indicator Code'] = df_indicateurs_reduit.index
print(f"Nombre d'indicateurs restants : {df_indicateurs_reduit.shape[0]}")

Nombre d'indicateurs restants : 16


df_indicateurs_reduit


# Création d'un dataframe de recherche
df_search = serie[['Series Code', 'Indicator Name','Topic','Long definition']]


# Target 15-24 ans
liste_1524 = []
for x in df_search['Series Code']:
    if "1524" in x:
        liste_1524.append(x)
df_1524 = df_search[df_search['Series Code'].isin(liste_1524)][['Series Code','Indicator Name','Topic','Long definition']]    
# On affiche la selection
pd.set_option('max_colwidth', None)
df_1524


# Target : Infrastructure Comprehensive list
liste_IT = [x for x in df_search["Long definition"] if "Internet" in x]

df_IT = df_search[df_search['Long definition'].isin(liste_IT)][['Series Code','Indicator Name','Topic','Long definition']]   
# On affiche la selection
pd.set_option('max_colwidth', None)
df_IT


# Target : Economie Comprehensive list
# first filter NY = incomplet ==> NY.GNP 
liste_ECO = [x for x in df_search["Series Code"] if ("NY.GNP") in x]

df_ECO = df_search[df_search['Series Code'].isin(liste_ECO)][['Series Code','Indicator Name','Topic','Long definition']]   
# On affiche la selection
pd.set_option('max_colwidth', None)
df_ECO


# Target : Etudiant engagés post Bac
# Deuxième filtre filtre
liste_Secondaire = [ x for x in df_search['Series Code'] if ("SEC.ENRL") in x]
        
df_Secondaire = df_search[df_search['Series Code'].isin(liste_Secondaire)][['Series Code','Indicator Name','Topic','Long definition']]   
# On affiche la selection
pd.set_option('max_colwidth', None)
df_Secondaire


# Target : Etudiant engagés post Bac
# Deuxième filtre filtre
liste_PostBac = [ x for x in df_search['Series Code'] if ("TER.ENR") in x]
        
df_PostBac = df_search[df_search['Series Code'].isin(liste_PostBac)][['Series Code','Indicator Name','Topic','Long definition']]   
# On affiche la selection
pd.set_option('max_colwidth', None)
df_PostBac


# Taux de remplissage sur 2010-2015
indicateurs_retenus=['SP.POP.1524.TO.UN', 'SE.SEC.ENRR', 'SE.TER.ENRR', 'IT.NET.USER.P2', 'NY.GNP.PCAP.PP.CD']
mask=df_indicateurs_reduit['Indicator Code'].isin(indicateurs_retenus)
cols=['Indicator Code','%']
df_indicateurs_reduit[mask][cols]


indicateurs_retenus=['IT.NET.USER.P2', 'NY.GNP.PCAP.PP.CD','SP.POP.1524.TO.UN', 'SE.SEC.ENRR', 'SE.TER.ENRR']


# On travail sur df
df = data
col = df.columns


# On supprime tout ce qui n'est pas pays et les pays na
print(liste_zones_et_income)

['ARB', 'EAS', 'EAP', 'EMU', 'ECS', 'ECA', 'EUU', 'HPC', 'HIC', 'LCN', 'LAC', 'LDC', 'LMY', 'LIC', 'LMC', 'MEA', 'MNA', 'MIC', 'NAC', 'OED', 'SAS', 'SSF', 'SSA', 'UMC', 'WLD', 'VGB', 'CUW', 'NRU', 'SXM', 'SSD', 'MAF']


df.shape

(886930, 86)


df = df[~df['Country Code'].isin(liste_zones_et_income)][col]
df.shape

(773315, 86)


df.columns

Index(['Country Name', 'Country Code', 'Indicator Name', 'Indicator Code',
       '1970', '1971', '1972', '1973', '1974', '1975', '1976', '1977', '1978',
       '1979', '1980', '1981', '1982', '1983', '1984', '1985', '1986', '1987',
       '1988', '1989', '1990', '1991', '1992', '1993', '1994', '1995', '1996',
       '1997', '1998', '1999', '2000', '2001', '2002', '2003', '2004', '2005',
       '2006', '2007', '2008', '2009', '2010', '2011', '2012', '2013', '2014',
       '2015', '2016', '2017', '2020', '2025', '2030', '2035', '2040', '2045',
       '2050', '2055', '2060', '2065', '2070', '2075', '2080', '2085', '2090',
       '2095', '2100', 'Unnamed: 69', '1970s', '1980s', '1990s', '2000s',
       '2010s', '2020s', '2030s', '2040s', '2050s', '2060s', '2070s', '2080s',
       '2090s', '2100s', 'Region', 'Income Group'],
      dtype='object')


# On réduit le df aux colonnes et dates (2000-2015) qui nous concernent
df = df.loc[:,[ 'Country Name', 'Country Code','Region', 'Indicator Name','Indicator Code','Income Group','2010', '2011','2012', '2013', '2014', '2015']]
df.shape

(773315, 12)


col = df.columns
print(col)

Index(['Country Name', 'Country Code', 'Region', 'Indicator Name',
       'Indicator Code', 'Income Group', '2010', '2011', '2012', '2013',
       '2014', '2015'],
      dtype='object')


# On supprime les indicateurs inutiles
df = df[df['Indicator Code'].isin(indicateurs_retenus)][col]   
df.shape

(1055, 12)


msno.bar(df)

<AxesSubplot: >


df[df['Indicator Code'].isin(indicateurs_retenus)][col].groupby(['Indicator Name', 'Indicator Code']).count().reset_index().sort_values(by='Indicator Name',ascending=False)


df.isnull().mean()*100

Country Name       0.000000
Country Code       0.000000
Region             0.473934
Indicator Name     0.000000
Indicator Code     0.000000
Income Group       0.473934
2010              19.620853
2011              18.672986
2012              19.810427
2013              21.895735
2014              22.559242
2015              26.255924
dtype: float64


# Df des lignes qui ont toutes des données manquantes tous les ans
df_noData = df.loc[(df['2010'].isnull()) & (df['2011'].isnull()) & (df['2012'].isnull()) & (df['2013'].isnull()) & (df['2014'].isnull()) & (df['2015'].isnull())]


pd.set_option('display.max_rows', None)
# On visualise les pays et les indicateurs concernés par le manque total de données
df_noData.loc[:,['Country Name','Country Code','Indicator Name']]


# faire la liste des codes pays a supprimer
liste_noData = df_noData['Country Code'].unique()
liste_noData

array(['ASM', 'AND', 'ATG', 'ABW', 'AZE', 'BHS', 'BMU', 'BOL', 'BIH',
       'BWA', 'KHM', 'CAN', 'CYM', 'CHI', 'CUB', 'DJI', 'DMA', 'GNQ',
       'FRO', 'FJI', 'PYF', 'GAB', 'GIB', 'GRL', 'GUM', 'GNB', 'HTI',
       'IRQ', 'IMN', 'KEN', 'KIR', 'PRK', 'XKX', 'LBY', 'LIE', 'MDV',
       'MHL', 'FSM', 'MCO', 'MNE', 'NAM', 'NCL', 'NIC', 'MNP', 'OMN',
       'PLW', 'PNG', 'PRI', 'WSM', 'SMR', 'SRB', 'SYC', 'SLE', 'SGP',
       'SLB', 'SOM', 'KNA', 'VCT', 'SUR', 'SYR', 'TON', 'TTO', 'TCA',
       'TUV', 'ARE', 'VUT', 'VEN', 'VNM', 'VIR', 'PSE', 'ZMB'],
      dtype=object)


df.shape

(1055, 12)


# On supprimes ces pays de notre dataframe
col = df.columns
df = df[~df['Country Code'].isin(liste_noData)][col]   
df.shape

(700, 12)


# On crée une nouvelle colonne last_data avec les valeurs de 2015
df['Last_data'] = df['2015']


df.head()


# Réindexing
df = df.reset_index()
df = df.drop('index',axis=1)
df.head()


#On parcours toute les lignes et on liste l'origine des données
nb_nan_LData = 0
nb_data_recup_2014 = 0
nb_data_recup_2013 = 0
nb_data_recup_2012 = 0
nb_data_recup_2011 = 0
nb_data_recup_2010 = 0

for i in range(len(df)):
    # Tester si nan  
    if(pd.isnull(df.loc[i,'Last_data'])):
        nb_nan_LData +=1
        if(pd.isnull(df.loc[i,'Last_data']))&(pd.notnull(df.loc[i,'2014'])):
            nb_data_recup_2014 +=1
        if(pd.isnull(df.loc[i,'Last_data']))&(pd.isnull(df.loc[i,'2014']))&(pd.notnull(df.loc[i,'2013'])):
            nb_data_recup_2013 +=1
        if(pd.isnull(df.loc[i,'Last_data']))&(pd.isnull(df.loc[i,'2014']))&(pd.isnull(df.loc[i,'2013']))&(pd.notnull(df.loc[i,'2012'])):
            nb_data_recup_2012 +=1
        if(pd.isnull(df.loc[i,'Last_data']))&(pd.isnull(df.loc[i,'2014']))&(pd.isnull(df.loc[i,'2013']))&(pd.isnull(df.loc[i,'2012']))&(pd.notnull(df.loc[i,'2011'])):
            nb_data_recup_2011 +=1
        if(pd.isnull(df.loc[i,'Last_data']))&(pd.isnull(df.loc[i,'2014']))&(pd.isnull(df.loc[i,'2013']))&(pd.isnull(df.loc[i,'2012']))&(pd.isnull(df.loc[i,'2011']))&(pd.notnull(df.loc[i,'2010'])):
            nb_data_recup_2010 +=1
nb_restant = nb_nan_LData - (nb_data_recup_2014 + nb_data_recup_2013  + nb_data_recup_2012 + nb_data_recup_2011 + nb_data_recup_2010)
print("Last_Data données manquantes : ",nb_nan_LData)
print("----------------------------------------------")
print("Origines des dernières données : ")
print("2014 : ",nb_data_recup_2014) 
print("2013 : ",nb_data_recup_2013)
print("2012 : ",nb_data_recup_2012)
print("2011 : ",nb_data_recup_2011)
print("2010 : ",nb_data_recup_2010)
print("restant : ",nb_restant)

Last_Data données manquantes :  102
----------------------------------------------
Origines des dernières données : 
2014 :  55
2013 :  20
2012 :  15
2011 :  8
2010 :  4
restant :  0


# On comble avec les dernières valeurs mais l'analyse sera biasée surtout si on fait de la prospective 
for i in range(len(df)):
    # Tester si nan  
    if(pd.isnull(df.loc[i,'Last_data'])):
        nb_nan_LData +=1
        if(pd.isnull(df.loc[i,'Last_data']))&(pd.notnull(df.loc[i,'2014'])):
            df.loc[i,'Last_data'] = df.loc[i,'2014']
            
        if(pd.isnull(df.loc[i,'Last_data']))&(pd.isnull(df.loc[i,'2014']))&(pd.notnull(df.loc[i,'2013'])):
            df.loc[i,'Last_data'] = df.loc[i,'2013']
        if(pd.isnull(df.loc[i,'Last_data']))&(pd.isnull(df.loc[i,'2014']))&(pd.isnull(df.loc[i,'2013']))&(pd.notnull(df.loc[i,'2012'])):
            df.loc[i,'Last_data'] = df.loc[i,'2012']
        if(pd.isnull(df.loc[i,'Last_data']))&(pd.isnull(df.loc[i,'2014']))&(pd.isnull(df.loc[i,'2013']))&(pd.isnull(df.loc[i,'2012']))&(pd.notnull(df.loc[i,'2011'])):
            df.loc[i,'Last_data'] = df.loc[i,'2011']
        if(pd.isnull(df.loc[i,'Last_data']))&(pd.isnull(df.loc[i,'2014']))&(pd.isnull(df.loc[i,'2013']))&(pd.isnull(df.loc[i,'2012']))&(pd.isnull(df.loc[i,'2011']))&(pd.notnull(df.loc[i,'2010'])):
            df.loc[i,'Last_data'] = df.loc[i,'2010']


# La colonne Last_data est complete
df.Last_data.isnull().mean()*100

0.0


print(f"{indicateurs_retenus}")

['IT.NET.USER.P2', 'NY.GNP.PCAP.PP.CD', 'SP.POP.1524.TO.UN', 'SE.SEC.ENRR', 'SE.TER.ENRR']


#On crée 1 df par indicateur
df_IT  = df[df['Indicator Code'] == 'IT.NET.USER.P2']
df_ECO = df[df['Indicator Code'] == 'NY.GNP.PCAP.PP.CD']
df_POP = df[df['Indicator Code'] == 'SP.POP.1524.TO.UN']
df_SEC = df[df['Indicator Code'] == 'SE.SEC.ENRR']
df_TER = df[df['Indicator Code'] == 'SE.TER.ENRR']
col = ['Country Name','Country Code','Region','Indicator Name','Indicator Code','Income Group','Last_data']
# On ne garde en date que last_data
df_IT  = df_IT.loc[:,col]
df_ECO = df_ECO.loc[:,col]
df_POP = df_POP.loc[:,col]
df_SEC = df_SEC.loc[:,col]
df_TER = df_TER.loc[:,col]

# Representation graphique des outliers:

fig = plt.figure(figsize=(20,20))

plt.subplot(5,2,1) # maillage des subplot
# Internet users (per 100 people)
sns.boxplot(data = df_IT, x='Indicator Code', y='Last_data',width=0.5, color='SteelBlue',showmeans=True,meanprops={"marker":"o",
                       "markerfacecolor":"red", 
                       "markeredgecolor":"black",
                      "markersize":"5"})
plt.title('Utilisateurs internet',fontweight='bold',size=15)
plt.ylabel('Nombre utilisateur internet en %')

plt.subplot(5,2,2)
sns.histplot(df_IT['Last_data'], kde = True, color='SteelBlue')
plt.title('Internet users (per 100 people)',fontweight='bold',size=15)
plt.xlabel('IT.NET.USER.P2')
plt.grid(False)

plt.subplot(5,2,3) # maillage des subplot
# GNI per capita, PPP (current international $)
sns.boxplot(data = df_ECO, x='Indicator Code', y='Last_data', width=0.5, color='SteelBlue',showmeans=True,meanprops={"marker":"o",
                       "markerfacecolor":"red", 
                       "markeredgecolor":"black",
                      "markersize":"5"})
plt.title('Economie',fontweight='bold',size=15)
plt.ylabel('GNI per capita, PPP (current international $)')

plt.subplot(5,2,4)
sns.histplot(df_ECO['Last_data'], kde = True, color='SteelBlue')
plt.title('GNI per capita, PPP (current international $)',fontweight='bold',size=15)
plt.xlabel('NY.GNP.PCAP.PP.CD')
plt.grid(False)

plt.subplot(5,2,5) # maillage des subplot
# Population, ages 15-24, total	
sns.boxplot(data = df_POP, x='Indicator Code', y='Last_data', width=0.5, color='SteelBlue',showmeans=True,meanprops={"marker":"o",
                       "markerfacecolor":"red", 
                       "markeredgecolor":"black",
                      "markersize":"7"})
plt.title('Population 15-24 ans',fontweight='bold',size=15)
plt.ylabel('Nombre de 15-24 ans')

plt.subplot(5,2,6)
sns.histplot(df_POP['Last_data'], kde = True, color='SteelBlue')
plt.title('Population, ages 15-24, total',fontweight='bold',size=15)
plt.xlabel('SP.POP.1524.TO.UN')
plt.grid(False)

plt.subplot(5,2,7) # maillage des subplot
# Enrolment in secondary education, both sexes (number)	
sns.boxplot(data = df_SEC, x='Indicator Code', y='Last_data', width=0.5, color='SteelBlue',showmeans=True,meanprops={"marker":"o",
                       "markerfacecolor":"red", 
                       "markeredgecolor":"black",
                      "markersize":"5"})
plt.title('Secondaire',fontweight='bold',size=15)
plt.ylabel('Nombre population dans le secondaire')

plt.subplot(5,2,8)
sns.histplot(df_SEC['Last_data'], kde = True, color='SteelBlue')
plt.title('Gross enrolment ratio, seconddary, both sexes (%)',fontweight='bold',size=15)
plt.xlabel('SE.SEC.ENRR')
plt.grid(False)

plt.subplot(5,2,9) # maillage des subplot
# Enrolment in tertiary education, all programmes, both sexes (number)	
sns.boxplot(data = df_TER, x='Indicator Code', y='Last_data', width=0.5, color='SteelBlue',showmeans=True,meanprops={"marker":"o",
                       "markerfacecolor":"red", 
                       "markeredgecolor":"black",
                      "markersize":"5"})
plt.title('Post Bac',fontweight='bold',size=15)
plt.ylabel('Nombre population en Post-Bac')

plt.subplot(5,2,10)
sns.histplot(df_TER['Last_data'], kde = True, color='SteelBlue')
plt.title('Gross enrolment ratio, tertiary, both sexes (%)',fontweight='bold',size=15)
plt.xlabel('SE.TER.ENRR')
plt.grid(False)

plt.subplots_adjust(left=0.125, # gerer les espacements
                    bottom=0.1, 
                    right=0.9, 
                    top=0.9, 
                    wspace=0.4, 
                    hspace=0.4)
plt.savefig("assets/graphiques/PREPROCESSING_2_Mesures distributions-1.jpg", bbox_inches='tight')
plt.show()


# Mesures de tendcances centrales
# Moyennes
mean_IT =df_IT['Last_data'].mean()
mean_ECO=df_ECO['Last_data'].mean()
mean_SEC=df_SEC['Last_data'].mean()
mean_TER=df_TER['Last_data'].mean()
mean_POP=df_POP['Last_data'].mean()

# Médianes
median_IT =df_IT['Last_data'].median()
median_ECO=df_ECO['Last_data'].median()
median_SEC=df_SEC['Last_data'].median()
median_TER=df_TER['Last_data'].median()
median_POP=df_POP['Last_data'].median()

# Mode 
# mode_IT =df_IT['Last_data'].mode()
# mode_ECO=df_ECO['Last_data'].mode()
# mode_SEC=df_SEC['Last_data'].mode()
# mode_TER=df_TER['Last_data'].mode()
# mode_POP=df_POP['Last_data'].mode()

# Variance
var_IT =df_IT['Last_data'].var(ddof=0)
var_ECO=df_ECO['Last_data'].var(ddof=0)
var_SEC=df_SEC['Last_data'].var(ddof=0)
var_TER=df_TER['Last_data'].var(ddof=0)
var_POP=df_POP['Last_data'].var(ddof=0)

# Ecart type
std_IT =df_IT['Last_data'].std(ddof=0)
std_ECO=df_ECO['Last_data'].std(ddof=0)
std_SEC=df_SEC['Last_data'].std(ddof=0)
std_TER=df_TER['Last_data'].std(ddof=0)
std_POP=df_POP['Last_data'].std(ddof=0)

# Les données suivent une loi normal si : skew=0 et Kurtosis =3

# skew : coefficient d'assymétrie (0 => symétrique)
skew_IT =df_IT['Last_data'].skew()
skew_ECO=df_ECO['Last_data'].skew()
skew_SEC=df_SEC['Last_data'].skew()
skew_TER=df_TER['Last_data'].skew()
skew_POP=df_POP['Last_data'].skew()

# Kortosis : coefficient d'applatissement proche de 3 suit loi normale
kurtosis_IT =df_IT['Last_data'].kurtosis()
kurtosis_ECO=df_ECO['Last_data'].kurtosis()
kurtosis_SEC=df_SEC['Last_data'].kurtosis()
kurtosis_TER=df_TER['Last_data'].kurtosis()
kurtosis_POP=df_POP['Last_data'].kurtosis()

# Visualisation des indicateurs
# On crée une liste des tendances centrales
stats = [['mean',mean_IT, mean_ECO, mean_SEC, mean_TER, mean_POP],
         ['median', median_IT, median_ECO, median_SEC,median_TER,median_POP], 
         ['var',var_IT, var_ECO, var_SEC,var_TER,var_POP], 
         ['std',std_IT, std_ECO,std_SEC, std_TER, std_POP],
         ['skew', skew_IT, skew_ECO, skew_SEC, skew_TER, skew_POP], 
         ['kurtosis', kurtosis_IT, kurtosis_ECO, kurtosis_SEC, kurtosis_TER, kurtosis_POP]]
# On crée un dataframe et des colonnes
df_stats = pd.DataFrame(stats,columns=['Statistique','IT_last_data','ECO_last_data','SEC_last_data','TER_last_data','POP_last_data'])
df_stats


# Création df avec tous les indicateurs en colonne
df_indicateurs_retenus = df.pivot_table(index=['Country Name'],columns='Indicator Code')['Last_data']


df_indicateurs_retenus.head()


df_indicateurs_retenus.describe()


# Identifier les outlier ECONOMIE
df_ECO.loc[df_ECO['Last_data']>60000]


# Identifier les outliers POP
df_POP.loc[df_POP['Last_data']>mean_POP].sort_values(ascending=False,by='Last_data')


# Identifier les outlier SEC
df_SEC.loc[df_SEC['Last_data']>150]


df_SEC_100 = df[df['Indicator Code'] == 'SE.SEC.ENRR']
df_TER_100 = df[df['Indicator Code'] == 'SE.TER.ENRR']
# On mets toutes les valeurs >100% a 100
df_SEC_100.Last_data[df_SEC_100.Last_data>100] = 100
df_TER_100.Last_data[df_TER_100.Last_data>100] = 100

# Representation graphique 

fig = plt.figure(figsize=(20,20))

plt.subplot(5,2,1) # maillage des subplot
# Internet users (per 100 people)
sns.boxplot(data = df_IT, x='Indicator Code', y='Last_data',width=0.5, color='SteelBlue',showmeans=True,meanprops={"marker":"o",
                                                                                                                   "markerfacecolor":"red",
                                                                                                                   "markeredgecolor":"black",
                                                                                                                   "markersize":"5"})
plt.title('Utilisateurs internet',fontweight='bold',size=15)
plt.ylabel('Nombre utilisateur internet en %')

plt.subplot(5,2,2)
sns.histplot(df_IT['Last_data'], kde = True, color='SteelBlue')
plt.title('Internet users (per 100 people)',fontweight='bold',size=15)
plt.xlabel('IT.NET.USER.P2')
plt.grid(False)

plt.subplot(5,2,3) # maillage des subplot
# GNI per capita, PPP (current international $)
sns.boxplot(data = df_ECO, x='Indicator Code', y='Last_data', width=0.5, color='SteelBlue',showmeans=True,meanprops={"marker":"o",
                       "markerfacecolor":"red", 
                       "markeredgecolor":"black",
                      "markersize":"5"})
plt.title('Economie',fontweight='bold',size=15)
plt.ylabel('GNI per capita, PPP (current international $)')

plt.subplot(5,2,4)
sns.histplot(df_ECO['Last_data'], kde = True, color='SteelBlue')
plt.title('GNI per capita, PPP (current international $)',fontweight='bold',size=15)
plt.xlabel('NY.GNP.PCAP.PP.CD')
plt.grid(False)

plt.subplot(5,2,5) # maillage des subplot
# Population, ages 15-24, total	
sns.boxplot(data = df_POP, x='Indicator Code', y='Last_data', width=0.5, color='SteelBlue',showmeans=True,meanprops={"marker":"o",
                                                                                                                     "markerfacecolor":"red", 
                                                                                                                     "markeredgecolor":"black",
                                                                                                                     "markersize":"7"})
plt.title('Population 15-24 ans',fontweight='bold',size=15)
plt.ylabel('Nombre de 15-24 ans')

plt.subplot(5,2,6)
sns.histplot(df_POP['Last_data'], kde = True, color='SteelBlue')
plt.title('Population, ages 15-24, total',fontweight='bold',size=15)
plt.xlabel('SP.POP.1524.TO.UN')
plt.grid(False)

plt.subplot(5,2,7) # maillage des subplot
# Enrolment in secondary education, both sexes (number)	
sns.boxplot(data = df_SEC_100, x='Indicator Code', y='Last_data', width=0.5, color='SteelBlue',showmeans=True,meanprops={"marker":"o",
                                                                                                                     "markerfacecolor":"red", 
                                                                                                                     "markeredgecolor":"black",
                                                                                                                     "markersize":"5"})
plt.title('Secondaire',fontweight='bold',size=15)
plt.ylabel('Nombre population dans le secondaire')

plt.subplot(5,2,8)
sns.histplot(df_SEC_100['Last_data'], kde = True, color='SteelBlue')
plt.title('Gross enrolment ratio, seconddary, both sexes (%)',fontweight='bold',size=15)
plt.xlabel('SE.SEC.ENRR')
plt.grid(False)

plt.subplot(5,2,9) # maillage des subplot
# Enrolment in tertiary education, all programmes, both sexes (number)	
sns.boxplot(data = df_TER_100, x='Indicator Code', y='Last_data', width=0.5, color='SteelBlue',showmeans=True,meanprops={"marker":"o",
                                                                                                                     "markerfacecolor":"red", 
                                                                                                                     "markeredgecolor":"black",
                                                                                                                     "markersize":"5"})
plt.title('Post Bac',fontweight='bold',size=15)
plt.ylabel('Nombre population en Post-Bac')

plt.subplot(5,2,10)
sns.histplot(df_TER_100['Last_data'], kde = True, color='SteelBlue')
plt.title('Gross enrolment ratio, tertiary, both sexes (%)',fontweight='bold',size=15)
plt.xlabel('SE.TER.ENRR')
plt.grid(False)

plt.subplots_adjust(left=0.125, # gerer les espacements
                    bottom=0.1, 
                    right=0.9, 
                    top=0.9, 
                    wspace=0.4, 
                    hspace=0.4)
plt.savefig("assets/graphiques/PREPROCESSING_3_Mesures distributions-2.jpg", bbox_inches='tight')
plt.show()


# Mesures de tendcances centrales
# Moyennes
mean_IT =df_IT['Last_data'].mean()
mean_ECO=df_ECO['Last_data'].mean()
mean_SEC_100=df_SEC_100['Last_data'].mean()
mean_TER_100=df_TER_100['Last_data'].mean()
mean_POP=df_POP['Last_data'].mean()

# Médianes
median_IT =df_IT['Last_data'].median()
median_ECO=df_ECO['Last_data'].median()
median_SEC_100=df_SEC_100['Last_data'].median()
median_TER_100=df_TER_100['Last_data'].median()
median_POP=df_POP['Last_data'].median()

# Mode 
# mode_IT =df_IT['Last_data'].mode()
# mode_ECO=df_ECO['Last_data'].mode()
# mode_SEC=df_SEC['Last_data'].mode()
# mode_TER=df_TER['Last_data'].mode()
# mode_POP=df_POP['Last_data'].mode()

# Variance
var_IT =df_IT['Last_data'].var(ddof=0)
var_ECO=df_ECO['Last_data'].var(ddof=0)
var_SEC_100=df_SEC_100['Last_data'].var(ddof=0)
var_TER_100=df_TER_100['Last_data'].var(ddof=0)
var_POP=df_POP['Last_data'].var(ddof=0)

# Ecart type
std_IT =df_IT['Last_data'].std(ddof=0)
std_ECO=df_ECO['Last_data'].std(ddof=0)
std_SEC_100=df_SEC_100['Last_data'].std(ddof=0)
std_TER_100=df_TER_100['Last_data'].std(ddof=0)
std_POP=df_POP['Last_data'].std(ddof=0)

# Les données suivent une loi normal si : skew=0 et Kurtosis =3

# skew : coefficient d'assymétrie (0 => symétrique)
skew_IT =df_IT['Last_data'].skew()
skew_ECO=df_ECO['Last_data'].skew()
skew_SEC_100=df_SEC_100['Last_data'].skew()
skew_TER_100=df_TER_100['Last_data'].skew()
skew_POP=df_POP['Last_data'].skew()

# Kortosis : coefficient d'applatissement proche de 3 suit loi normale
kurtosis_IT =df_IT['Last_data'].kurtosis()
kurtosis_ECO=df_ECO['Last_data'].kurtosis()
kurtosis_SEC_100=df_SEC_100['Last_data'].kurtosis()
kurtosis_TER_100=df_TER_100['Last_data'].kurtosis()
kurtosis_POP=df_POP['Last_data'].kurtosis()

# Visualisation des indicateurs
# On crée une liste des tendances centrales
stats = [['mean',mean_IT, mean_ECO, mean_SEC_100, mean_TER_100, mean_POP],
         ['median', median_IT, median_ECO, median_SEC_100,median_TER_100,median_POP], 
         ['var',var_IT, var_ECO, var_SEC_100,var_TER_100,var_POP], 
         ['std',std_IT, std_ECO,std_SEC_100, std_TER_100, std_POP],
         ['skew', skew_IT, skew_ECO, skew_SEC_100, skew_TER_100, skew_POP], 
         ['kurtosis', kurtosis_IT, kurtosis_ECO, kurtosis_SEC_100, kurtosis_TER_100, kurtosis_POP]]
# On crée un dataframe et des colonnes
df_stats_100 = pd.DataFrame(stats,columns=['Statistique','IT_last_data','ECO_last_data','SEC_last_data','TER_last_data','POP_last_data'])
df_stats_100


# On remet les indicateurs à 100%
df_indicateurs_retenus.loc[df_indicateurs_retenus['SE.SEC.ENRR']>100] = 100
df_indicateurs_retenus.loc[df_indicateurs_retenus['SE.TER.ENRR']>100] = 100


df_indicateurs_retenus.describe()


df_indicateurs_cleaned = df_indicateurs_retenus.copy()


df_indicateurs_cleaned.describe()


# Fonction de visualisation des indicateurs normalisés pour fair la selection de la transormation la plus adéquat
def graph_norm(df,titre):
    titre = "assets/graphiques/"+titre+".jpeg"
    plt.figure(figsize=(10,5))
    col=indicateurs_retenus
    colors=['orange','Green','sienna','SteelBlue','Purple']
    for i in range(5): 
        sns.kdeplot(df[col[i]],label=df.columns[i],color=colors[i])
    plt.legend()
    plt.grid(False)
    plt.xlabel('')
    # plt.savefig(titre)
    plt.show()


# On normalise avec scit-learb
# from sklearn import preprocessing
cols = ['IT.NET.USER.P2','NY.GNP.PCAP.PP.CD','SE.SEC.ENRR','SE.TER.ENRR','SP.POP.1524.TO.UN']


scaler = nltk.preprocessing.StandardScaler()
scaled = scaler.fit_transform(df_indicateurs_cleaned[cols])
df_StandardScaler = pd.DataFrame(scaled, columns=cols)
# df_StandardScaler.head()
graph_norm(df_StandardScaler,titre="NORMALISATION_1_StandardScaler_100")


df_norm = df_indicateurs_cleaned.copy()


# 'IT.NET.USER.P2', 'NY.GNP.PCAP.PP.CD', 'SP.POP.1524.TO.UN', 'SE.SEC.ENRR', 'SE.TER.ENRR'
indicateur = "NY.GNP.PCAP.PP.CD"
print(f"max : {df_norm.loc[:,[indicateur]].max()}")
print(f"min : {df_norm.loc[:,[indicateur]].min()}")
print(f"median : {df_norm.loc[:,[indicateur]].median()}")

max : Indicator Code
NY.GNP.PCAP.PP.CD    125000.0
dtype: float64
min : Indicator Code
NY.GNP.PCAP.PP.CD    100.0
dtype: float64
median : Indicator Code
NY.GNP.PCAP.PP.CD    3170.0
dtype: float64


# 'IT.NET.USER.P2', 'NY.GNP.PCAP.PP.CD', 'SP.POP.1524.TO.UN', 'SE.SEC.ENRR', 'SE.TER.ENRR'
indicateur = "SP.POP.1524.TO.UN"
print(f"max : {df_norm.loc[:,[indicateur]].max()}")
print(f"min : {df_norm.loc[:,[indicateur]].min()}")
print(f"median : {df_norm.loc[:,[indicateur]].median()}")

max : Indicator Code
SP.POP.1524.TO.UN    244120201.0
dtype: float64
min : Indicator Code
SP.POP.1524.TO.UN    100.0
dtype: float64
median : Indicator Code
SP.POP.1524.TO.UN    596897.5
dtype: float64


df_norm = df_norm.loc[df_norm['NY.GNP.PCAP.PP.CD']>5430.0]
df_norm = df_norm.loc[df_norm['SP.POP.1524.TO.UN']>596897.5]
df_norm.head()


scaler = nltk.preprocessing.StandardScaler()
scaled = scaler.fit_transform(df_norm[cols])
df_StandardScaler_norm = pd.DataFrame(scaled, columns=cols)
# df_StandardScaler_n.head()
graph_norm(df_StandardScaler_norm,titre="NORMALISATION_2_StandardScaler_norm")


scaler = nltk.preprocessing.PowerTransformer()
scaled = scaler.fit_transform(df_norm[cols])
df_PowerTransformer_norm = pd.DataFrame(scaled, columns=cols)
# df_StandardScaler_n.head()
graph_norm(df_PowerTransformer_norm,titre="NORMALISATION_3_PowerTransform_norm")


df_indicateurs_nonCleaner = df.pivot_table(index=['Country Name'],columns='Indicator Code')['Last_data']
scaler = nltk.preprocessing.PowerTransformer(method = 'box-cox')
df_PowerTransformer = scaler.fit_transform(df_indicateurs_nonCleaner[cols])
df_PowerTransformer = pd.DataFrame(df_PowerTransformer, columns=cols)
df_PowerTransformer.head()
graph_norm(df_PowerTransformer,titre="NORMALISATION_3_PowerTransform_non_clean_OK")


df_stats = df.copy()


# On supprime les anées 2010	2011	2012	2013	2014	2015
df_stats = df_stats.drop(['2010','2011','2012','2013','2014','2015'],axis=1)


#Oncrée les colone de base
a2 = pd.DataFrame({'Country Name' : df_stats['Country Name'],
                   'Country Code' : df_stats['Country Code'],
                   'Region' : df_stats['Region'],
                   'Income Group' : df_stats['Income Group']})


# On recupère des dernier données par indicateur
a3 = df_stats.pivot_table(index=['Country Code'],columns='Indicator Code')['Last_data']
a3.loc[a3['SE.SEC.ENRR']>100] = 100
a3.loc[a3['SE.TER.ENRR']>100] = 100


# On fusionne les deux pour avoir toutes les informations
df_stats = pd.merge(a2,a3,left_on="Country Code", right_on='Country Code', how='inner')
df_stats = df_stats.drop_duplicates()


df_stats.head()


# On affiche les ordres de grandeur des indicateurs
df_stats.describe()


df_stats_region = df_stats.copy()


df_stats_region.head()


for i in range(len(indicateurs_retenus)):
    print('--------------------------------------------------------')
    print(indicateurs_retenus[i])
    print('--------------------------------------------------------')
    print(df_stats_region.groupby('Region')[indicateurs_retenus[i]].describe(include=all))

--------------------------------------------------------
IT.NET.USER.P2
--------------------------------------------------------
                            count       mean        std        min        25%  \
Region                                                                          
East Asia & Pacific          16.0  63.740673  33.526813  18.200000  22.744017   
Europe & Central Asia        44.0  85.080086  24.608391  14.996775  70.235925   
Latin America & Caribbean    21.0  58.495446  25.891286  26.802605  41.590000   
Middle East & North Africa   14.0  66.553292  25.806430  24.085409  46.131191   
North America                 1.0  74.554202        NaN  74.554202  74.554202   
South Asia                    7.0  21.432908  10.988467   8.260000  14.200000   
Sub-Saharan Africa           37.0  16.669778  12.904877   1.083733   7.459161   

                                   50%         75%         max  
Region                                                          
East Asia & Pacific          71.132034  100.000000  100.000000  
Europe & Central Asia       100.000000  100.000000  100.000000  
Latin America & Caribbean    53.810000   64.600000  100.000000  
Middle East & North Africa   67.057191   89.044437  100.000000  
North America                74.554202   74.554202   74.554202  
South Asia                   17.581618   27.994369   39.800000  
Sub-Saharan Africa           12.400000   22.742818   51.919116  
--------------------------------------------------------
NY.GNP.PCAP.PP.CD
--------------------------------------------------------
                            count          mean           std      min  \
Region                                                                   
East Asia & Pacific          16.0  18923.750000  29947.449279    100.0   
Europe & Central Asia        44.0   4573.181818   8358.485668    100.0   
Latin America & Caribbean    21.0   9083.809524   6591.388682    100.0   
Middle East & North Africa   14.0  23414.285714  35963.664508    100.0   
North America                 1.0  57900.000000           NaN  57900.0   
South Asia                    7.0   5481.428571   3323.750034   1880.0   
Sub-Saharan Africa           37.0   3392.432432   3750.090670    670.0   

                                25%      50%      75%       max  
Region                                                           
East Asia & Pacific           100.0   7205.0  17300.0   98650.0  
Europe & Central Asia         100.0    100.0   5617.5   31430.0  
Latin America & Caribbean    4290.0   8370.0  13590.0   20570.0  
Middle East & North Africa   3942.5  10855.0  16780.0  125000.0  
North America               57900.0  57900.0  57900.0   57900.0  
South Asia                   3035.0   5320.0   6800.0   11500.0  
Sub-Saharan Africa           1500.0   1990.0   3700.0   19980.0  
--------------------------------------------------------
SP.POP.1524.TO.UN
--------------------------------------------------------
                            count          mean           std         min  \
Region                                                                      
East Asia & Pacific          16.0  1.757077e+07  4.909192e+07       100.0   
Europe & Central Asia        44.0  4.676263e+05  1.188618e+06       100.0   
Latin America & Caribbean    21.0  3.826213e+06  8.278347e+06       100.0   
Middle East & North Africa   14.0  3.716807e+06  5.170341e+06       100.0   
North America                 1.0  4.514752e+07           NaN  45147517.0   
South Asia                    7.0  4.782893e+07  8.793947e+07    147444.0   
Sub-Saharan Africa           37.0  4.873992e+06  6.976425e+06     38279.0   

                                   25%         50%          75%          max  
Region                                                                        
East Asia & Pacific              100.0    398884.5   7039125.00  197026759.0  
Europe & Central Asia            100.0       100.0    461100.75    5897703.0  
Latin America & Caribbean      23536.0    536826.0   1962188.00   33595574.0  
Middle East & North Africa     70972.5   1053006.5   6063574.75   15377746.0  
North America               45147517.0  45147517.0  45147517.00   45147517.0  
South Asia                   4900342.0   7252785.0  36740702.50  244120201.0  
Sub-Saharan Africa            978071.0   3009775.0   4945440.00   36659023.0  
--------------------------------------------------------
SE.SEC.ENRR
--------------------------------------------------------
                            count       mean        std        min        25%  \
Region                                                                          
East Asia & Pacific          16.0  88.672663  14.829773  51.296089  83.774380   
Europe & Central Asia        44.0  97.535537   4.987859  79.294098  98.816381   
Latin America & Caribbean    21.0  88.640386  11.284546  65.626701  79.425407   
Middle East & North Africa   14.0  86.115675  15.910126  48.624660  83.359735   
North America                 1.0  97.564690        NaN  97.564690  97.564690   
South Asia                    7.0  69.822295  18.283007  44.525871  59.584469   
Sub-Saharan Africa           37.0  47.683012  20.210481  17.378010  33.666660   

                                   50%         75%         max  
Region                                                          
East Asia & Pacific          95.189339  100.000000  100.000000  
Europe & Central Asia       100.000000  100.000000  100.000000  
Latin America & Caribbean    90.549690   99.650970  100.000000  
Middle East & North Africa   90.169865   98.638306  100.000000  
North America                97.564690   97.564690   97.564690  
South Asia                   67.169853   79.083447   99.724510  
Sub-Saharan Africa           43.395191   56.811958   98.819672  
--------------------------------------------------------
SE.TER.ENRR
--------------------------------------------------------
                            count       mean        std        min        25%  \
Region                                                                          
East Asia & Pacific          16.0  59.140481  35.885345  13.528170  25.619302   
Europe & Central Asia        44.0  83.347298  27.466859   7.984440  66.814675   
Latin America & Caribbean    21.0  52.388614  32.196095  12.480620  27.220270   
Middle East & North Africa   14.0  49.291286  31.134223   9.974600  29.759329   
North America                 1.0  85.795776        NaN  85.795776  85.795776   
South Asia                    7.0  14.938481   6.440884   8.662800  10.427160   
Sub-Saharan Africa           37.0   9.451695   6.853559   0.797730   4.775910   

                                   50%         75%         max  
Region                                                          
East Asia & Pacific          55.979393  100.000000  100.000000  
Europe & Central Asia       100.000000  100.000000  100.000000  
Latin America & Caribbean    40.512791   91.147743  100.000000  
Middle East & North Africa   37.703526   65.766324  100.000000  
North America                85.795776   85.795776   85.795776  
South Asia                   13.440800   17.368285   26.874880  
Sub-Saharan Africa            8.433270   10.849330   36.667568


fig = plt.figure(figsize=(20,20))
fig.suptitle(' Ordres de grandeurs par indicateur et par region (en moyenne)',fontweight='bold', fontsize=20)
plt.subplot(5,2,1) # maillage des subplot
plt.title('Accès internet : IT.NET.USER.P2',fontweight='bold',size=15)
df_stats_region.groupby('Region')['IT.NET.USER.P2'].mean().sort_values(ascending=True).plot(kind='barh',width=0.8)
plt.grid(False)

plt.subplot(5,2,2) # maillage des subplot
plt.title("Economie : NY.GNP.PCAP.PP.CD",fontweight='bold',size=15)
df_stats_region.groupby('Region')['NY.GNP.PCAP.PP.CD'].mean().sort_values(ascending=True).plot(kind='barh',width=0.8)
plt.grid(False)

plt.subplot(5,2,3) # maillage des subplot
plt.title("Population 15-20 ans : SP.POP.1524.TO.UN",fontweight='bold',size=15)
df_stats_region.groupby('Region')['SP.POP.1524.TO.UN'].mean().sort_values(ascending=True).plot(kind='barh',width=0.8)
plt.grid(False)

plt.subplot(5,2,4) # maillage des subplot
plt.title("Enseignement Secondaire : SE.SEC.ENRR",fontweight='bold',size=15)
df_stats_region.groupby('Region')['SE.SEC.ENRR'].mean().sort_values(ascending=True).plot(kind='barh',width=0.8)
plt.grid(False)

plt.subplot(5,2,5) # maillage des subplot
plt.title("Post Bac : SE.TER.ENRR",fontweight='bold',size=15)
df_stats_region.groupby('Region')['SE.TER.ENRR'].mean().sort_values(ascending=True).plot(kind='barh',width=0.8)
width=1
plt.grid(False)

plt.subplots_adjust(left=0.125, # gerer les espacements
                    bottom=0.1, 
                    right=0.9, 
                    top=0.93, 
                    wspace=0.4, 
                    hspace=0.4
                   )
plt.savefig("assets/graphiques/PREPROCESSING_4_Ordre de grandeurs.jpg", bbox_inches='tight')
plt.show()


df_score = df_PowerTransformer
df_score['Pays'] = df_indicateurs_retenus.index


# On calcule et affecte un scoring
for i in range(len(df_score)):
    df_score.loc[i,'Score'] = (round((df_score.loc[i,'IT.NET.USER.P2']*5) 
                             +(df_score.loc[i,'SP.POP.1524.TO.UN']*4)
                             + (df_score.loc[i,'SE.TER.ENRR']*3)
                             + (df_score.loc[i,'SE.SEC.ENRR']*2)
                             + (df_score.loc[i,'NY.GNP.PCAP.PP.CD']*1)/14,2)
                            )


# TOP 10
df_top_10 = df_score.sort_values(by='Score',ascending=False).head(10)
df_top_10


df_top_10 = df_top_10.reset_index()
df_top_10 = df_top_10.drop(['index'],axis=1)
liste_pays_retenus = df_top_10['Pays'].tolist()


df_score_norm = df_PowerTransformer_norm
df_score_norm['Pays'] = df_norm.index


for i in range(len(df_score_norm)):
    df_score_norm.loc[i,'Score'] = (round((df_score_norm.loc[i,'IT.NET.USER.P2']*5) 
                             +(df_score_norm.loc[i,'SP.POP.1524.TO.UN']*4)
                             + (df_score_norm.loc[i,'SE.TER.ENRR']*3)
                             + (df_score_norm.loc[i,'SE.SEC.ENRR']*2)
                             + (df_score_norm.loc[i,'NY.GNP.PCAP.PP.CD']*1)/14,2)
                            )   
# TOP 10
df_top_10_norm = df_score_norm.sort_values(by='Score',ascending=False).head(10)
df_top_10_norm


# On ajoute la region a df_score
df_score_region_sc = pd.merge(df_score,df_stats_region,left_on="Pays", right_on='Country Name', how='inner')
df_score_region_sc.columns

Index(['IT.NET.USER.P2_x', 'NY.GNP.PCAP.PP.CD_x', 'SE.SEC.ENRR_x',
       'SE.TER.ENRR_x', 'SP.POP.1524.TO.UN_x', 'Pays', 'Score', 'Country Name',
       'Country Code', 'Region', 'Income Group', 'IT.NET.USER.P2_y',
       'NY.GNP.PCAP.PP.CD_y', 'SE.SEC.ENRR_y', 'SE.TER.ENRR_y',
       'SP.POP.1524.TO.UN_y'],
      dtype='object')


# On supprime les colonne des indicateurs non normalisés
df_score_region_sc = df_score_region_sc.drop(columns=['IT.NET.USER.P2_y', 'Country Name','Country Code',
                                                      'NY.GNP.PCAP.PP.CD_y', 'SE.SEC.ENRR_y', 'SE.TER.ENRR_y',
                                                      'SP.POP.1524.TO.UN_y'],axis=1)


df_score_income_sc = df_score_region_sc.copy()


# On renomme les colonnes
df_score_region_sc = df_score_region_sc.rename(columns={'IT.NET.USER.P2_x':'NET.USER',
                                                        'NY.GNP.PCAP.PP.CD_x':'ECO',
                                                        'SE.SEC.ENRR_x':'SEC',
                                                        'SE.TER.ENRR_x':'TER',
                                                        'SP.POP.1524.TO.UN_x':'POP'})


# On regroupe par region et on fait la moyenne des indicateurs par region
df_score_region_sc = df_score_region_sc.groupby('Region').mean()
df_score_region_sc = df_score_region_sc.sort_values(by='Score',ascending=False)
df_score_region_sc


# Import Pi
from math import pi


# Set data
df_spider = df_score_region_sc[['NET.USER','ECO','SEC','TER','POP']]
df_spider


plt.rc('axes', facecolor = 'Gainsboro')

# Nombre de variables
categories=list(df_spider.columns)

# Nombre de catégorie = nombre de colonne = 5
N = len(categories)
 
# Pour angle on divise le plot par le nombre de variable-> 1/4 cercle -> *2 -> demi cercle * pi cercle entier
angles = [n / N*2*pi for n in range(N)]

# On fixe la taille de la figure
fig = plt.figure(1, figsize=(20,18))

# On choisit les colonnes
# colors = ['cadetblue', 'lightseagreen', 'darkseagreen', 'indianred', 'steelblue', 'RoyalBlue','lightcoral']
colors = ['seagreen','mediumseagreen','darkseagreen', 'lightseagreen', 'steelblue', 'coral','lightcoral']
# Boucle pour chacune des régions contenues dans l'index
for i, region in enumerate(df_spider.index) : 
    # On crée un sous plot
    ax = fig.add_subplot(3,3,i+1, polar=True)
    
    # If you want the first axis to be on top:
    ax.set_theta_offset(pi/2)
    ax.set_theta_direction(-1)

    # Draw one axe per variable + add labels labels yet
    plt.xticks(angles, categories, color='black', size=8)
    # On supprime la visibilité des graduations circulaires
    plt.yticks(color="gray", size=0.1)
    
    # On definit values = contenue
    values=df_spider.iloc[i].values
    
    # On dessine angle, values, motif, taille du trait, style du trait
    ax.plot(angles, values, '.',color='black', linewidth=0, linestyle='solid')
    # On definit la visualisation de l'aire dessiné
    ax.fill(angles, values, colors[i], alpha=0.6)
    
    # On aggrandit les labels 
    ax.set_xticklabels(categories, fontsize=12)
    # Titer des graphiques et couleurs
    ax.set_title(region, size=20, color=colors[i])
    # grille
    plt.grid(True,color="grey")
    # On definit l'étendu de y entre -2 et 2 pour une bonne visualisation
    plt.ylim(-2,2)
plt.savefig("assets/graphiques/PREPROCESSING_5_Ordre de grandeurs_Regions.jpg", bbox_inches='tight')


fig = plt.figure(1, figsize=(10,6))
res = sns.heatmap(df_spider, square=True, linewidths=0.1, cmap='coolwarm')
res.set_yticklabels(res.get_ymajorticklabels(), fontsize = 14)
res.set_xticklabels(res.get_xmajorticklabels(), fontsize = 12)
plt.title("Heatmap Indicateurs Region Score",fontsize=18,fontweight='bold')
# plt.savefig("assets/graphiques/PREPROCESSING_6_Heatmap_indicateurs_region.jpg")
plt.show()


df_score_income_sc.columns

Index(['IT.NET.USER.P2_x', 'NY.GNP.PCAP.PP.CD_x', 'SE.SEC.ENRR_x',
       'SE.TER.ENRR_x', 'SP.POP.1524.TO.UN_x', 'Pays', 'Score', 'Region',
       'Income Group'],
      dtype='object')


df_score_income_sc=df_score_income_sc.groupby('Income Group').mean()


# On renomme les colonnes
df_score_income_sc = df_score_income_sc.rename(columns={'IT.NET.USER.P2_x':'NET.USER',
                                                        'NY.GNP.PCAP.PP.CD_x':'ECO',
                                                        'SE.SEC.ENRR_x':'SEC',
                                                        'SE.TER.ENRR_x':'TER',
                                                        'SP.POP.1524.TO.UN_x':'POP'})


df_score_income_sc.head()


df_spider2 = df_score_income_sc[['NET.USER','ECO','SEC','TER','POP']]
plt.rc('axes', facecolor = 'Gainsboro')

# Nombre de variables
categories=list(df_spider2.columns)

# Nombre de catégorie = nombre de colonne = 5
N = len(categories)
 
# Pour angle on divise le plot par le nombre de variable-> 1/4 cercle -> *2 -> demi cercle * pi cercle entier
angles = [n / N*2*pi for n in range(N)]

# On fixe la taille de la figure
fig = plt.figure(1, figsize=(20,18))

# On choisit les colonnes
# colors = ['sienna', 'LightGreen', 'steelblue', 'RoyalBlue','Peru']
colors = ['seagreen','mediumseagreen','darkseagreen', 'lightseagreen', 'salmon']

# Boucle pour chacune des régions contenues dans l'index
for i, region in enumerate(df_spider2.index) : 
    # On crée un sous plot
    ax = fig.add_subplot(3,3,i+1, polar=True)
    
    # If you want the first axis to be on top:
    ax.set_theta_offset(pi/2)
    ax.set_theta_direction(-1)

    # Draw one axe per variable + add labels labels yet
    plt.xticks(angles, categories, color='black', size=8)
    # On supprime la visibilité des graduations circulaires
    plt.yticks(color="gray", size=0.1)
    
    # On definit values = contenue
    values=df_spider.iloc[i].values
    
    # On dessine angle, values, motif, taille du trait, style du trait
    ax.plot(angles, values, '.',color='black', linewidth=0, linestyle='solid')
    # On definit la visualisation de l'aire dessiné
    ax.fill(angles, values, colors[i], alpha=0.6)
    
    # On aggrandit les labels 
    ax.set_xticklabels(categories, fontsize=12)
    # Titer des graphiques et couleurs
    ax.set_title(region, size=20, color=colors[i])
    # grille
    plt.grid(True,color="grey")
    # On definit l'étendu de y entre -2 et 2 pour une bonne visualisation
    plt.ylim(-2,2)
plt.savefig("assets/graphiques/PREPROCESSING_7_Ordre de grandeurs_Income.jpg", bbox_inches='tight')


df_spider2 = df_score_income_sc[['NET.USER','ECO','SEC','TER','POP']]
fig = plt.figure(1, figsize=(10,6))
res = sns.heatmap(df_spider2, square=True, linewidths=0.1, cmap='coolwarm')
res.set_yticklabels(res.get_ymajorticklabels(), fontsize = 14)
res.set_xticklabels(res.get_xmajorticklabels(), fontsize = 12)
plt.title("heatmap Indicateurs Income Score",fontsize=18,fontweight='bold')
# plt.savefig("assets/graphiques/PREPROCESSING_8_Heatmap indicateur Income.jpg")
plt.show()


# PANDAS
#Import des pays du monde de geopandas
world = gpd.read_file(gpd.datasets.get_path('naturalearth_lowres')) 
# Préparation des données
# Map_1 : préparer les données côté geopandas
map_1 = world.sort_values(by=['iso_a3'])
map_1 = map_1.loc[:,['name','iso_a3']]
map_1 = map_1.rename(columns={'iso_a3':'iso3'})
map_1 = map_1.set_index('iso3')  
# Map 2 : préparer les données côté EdStatsCountry
map_2=country.sort_values(by = ['Country Code'])
map_2=map_2.loc[:,['Country Code','Short Name']]
map_2=map_2.rename(columns={'Country Code':'iso3'})
map_2=map_2.set_index('iso3')
# Fusion des 2 dataframes
map_f = pd.merge(map_1,map_2, on='iso3', how='left')
map_f.reset_index(inplace=True)

# Recherche des noms de pays différents entre geopandas et notre jeu de données
rows = map_f[map_f['name'] != map_f['Short Name']]
rows.drop([0,7,8,39,50,56,144,163], inplace=True)
# Modifier World
df_comp=rows[rows['Short Name'].notnull()].sort_values(by='iso3')
noms_a_changer=df_comp['Short Name'].tolist()
codes_iso=df_comp['iso3'].tolist()
# Récupération des index geopandas des noms de pays à changer
index_geo=world[world['iso_a3'].isin(codes_iso)].index
# Récupération des codes ISO 3 de geopandas
codes_isoa3=world.loc[index_geo,'iso_a3'].tolist()
# Parcours les codes iso3 pour avoir une liste de noms de pays dans l'ordre de world
noms_a_changer=[]
[noms_a_changer.append(df_comp[df_comp['iso3'] == iso]['Short Name'].tolist()[0]) for iso in codes_isoa3]
# On remplace les noms de world par ceux de notre dataframe
world.loc[world.index[index_geo.tolist()],'name']=noms_a_changer

df_score.Pays[df_score.Pays=='Korea, Rep.']="Korea"
df_score.Pays[df_score.Pays=='Russian Federation'] = 'Russia'
df_score_geo = world.merge(df_score, left_on = 'name', right_on = 'Pays')[['geometry','Pays','Score']]


# import matplotlib.pyplot as plt
from mpl_toolkits.axes_grid1.axes_divider import make_axes_locatable
fig, ax = plt.subplots()
fig.set_size_inches(30,18)

# créer sous division des axes
divider = make_axes_locatable(ax)
cax = divider.append_axes("right", size="2%", pad=0.1)
# graph
world.plot(legend = True,ax=ax, cax=cax, color='lightgrey',linestyle='-',edgecolor='black')
df_score_geo.plot(column=df_score_geo['Score'], legend = True, ax=ax, cax=cax, cmap='RdYlGn',linestyle="-",edgecolor='black')
ax.set_axis_off()
# plt.savefig("assets/graphiques/_MAP_1_Pays_indicateurs.jpg")
plt.show()


df_prediction = data.copy()


df_prediction = data[data['Country Name'].isin(liste_pays_retenus) & data['Indicator Code'].isin(indicateurs_retenus)]

df_prediction = df_prediction.melt(id_vars = ['Country Name', 'Country Code', 'Indicator Name', 'Indicator Code'], 
                                   value_vars = ['1970', '1971', '1972', '1973', '1974', '1975', '1976', '1977', '1978',
                                                 '1979', '1980', '1981', '1982', '1983', '1984', '1985', '1986', '1987',
                                                 '1988', '1989', '1990', '1991', '1992', '1993', '1994', '1995', '1996',
                                                 '1997', '1998', '1999', '2000', '2001', '2002', '2003', '2004', '2005',
                                                 '2006', '2007', '2008', '2009', '2010', '2011', '2012', '2013', '2014',
                                                 '2015', '2016', '2017', '2020', '2025', '2030', '2035', '2040', '2045',
                                                 '2050', '2055', '2060', '2065', '2070', '2075', '2080', '2085', '2090',
                                                 '2095', '2100'], var_name = 'Year', value_name = 'Value')

df_prediction['Year'] = df_prediction['Year'].astype('int32')
df_prediction = df_prediction[df_prediction['Year'] > 2010]


# Representation graphique

fig = plt.figure(figsize=(20,20))
fig.suptitle("Evolution des indicateurs des pays sélectionner en fonction des années disponible dans dataEdStats",fontweight='bold', fontsize=20)
plt.subplot(5,2,1) 
for i in liste_pays_retenus:
    sns.lineplot(
        x=df_prediction[df_prediction['Indicator Code'] == 'IT.NET.USER.P2'][df_prediction['Country Name'] == i ]['Year'],
        y=df_prediction[df_prediction['Indicator Code'] == 'IT.NET.USER.P2'][df_prediction['Country Name'] == i ]['Value'],
        label = i,
        linewidth=2.5,
        legend=False,
    )
    
#plt.legend(bbox_to_anchor=(1, 1), loc = 2)
plt.title("NUMERIQUE", size=20,weight='bold')
plt.ylabel("%",rotation=0)
plt.xlabel("Années")
plt.grid(False)

plt.subplot(5,2,2)
for i in liste_pays_retenus:
    sns.lineplot(
        x=df_prediction[df_prediction['Indicator Code'] == 'NY.GNP.PCAP.PP.CD'][df_prediction['Country Name'] == i ]['Year'],
        y=df_prediction[df_prediction['Indicator Code'] == 'NY.GNP.PCAP.PP.CD'][df_prediction['Country Name'] == i ]['Value'],
        label = i,
        linewidth=2.5,
        legend=False
    )
#plt.legend(bbox_to_anchor=(1, 1), loc = 2)
plt.title("ECONOMIE", size=20,weight='bold')
plt.ylabel("",rotation=0)
plt.xlabel("Années")
plt.grid(False)

plt.subplot(5,2,3) 
for i in liste_pays_retenus:
    sns.lineplot(
        x=df_prediction[df_prediction['Indicator Code'] == 'SP.POP.1524.TO.UN'][df_prediction['Country Name'] == i ]['Year'],
        y=df_prediction[df_prediction['Indicator Code'] == 'SP.POP.1524.TO.UN'][df_prediction['Country Name'] == i ]['Value'],
        label = i,
        linewidth=2.5,
        legend=False
    )
    
#plt.legend(bbox_to_anchor=(1, 1), loc = 2)
plt.title("POPULATION", size=20,weight='bold')
plt.ylabel("",rotation=0)
plt.xlabel("Années")
plt.grid(False)

plt.subplot(5,2,4)
for i in liste_pays_retenus:
    sns.lineplot(
        x=df_prediction[df_prediction['Indicator Code'] == 'SE.SEC.ENRR'][df_prediction['Country Name'] == i ]['Year'],
        y=df_prediction[df_prediction['Indicator Code'] == 'SE.SEC.ENRR'][df_prediction['Country Name'] == i ]['Value'],
        label = i,
        linewidth=2.5,
        legend=False
    )
    
# plt.legend(bbox_to_anchor=(1, 1), loc = 2)
plt.title("ETUDES SECONDAIRES", size=20,weight='bold')
plt.ylabel("%",rotation=0)
plt.xlabel("Années")
plt.grid(False)

plt.subplot(5,2,5) 
for i in liste_pays_retenus:
    sns.lineplot(
        x=df_prediction[df_prediction['Indicator Code'] == 'SE.TER.ENRR'][df_prediction['Country Name'] == i ]['Year'],
        y=df_prediction[df_prediction['Indicator Code'] == 'SE.TER.ENRR'][df_prediction['Country Name'] == i ]['Value'],
        label = i,
        linewidth=2.5
    )
plt.legend(bbox_to_anchor=(1.5, 1), loc = 2)
plt.title("ETUDES SUPERIEURES", size=20,weight='bold')
plt.ylabel("%",rotation=0)
plt.xlabel("Années")
plt.grid(False)	


plt.subplots_adjust(left=0.125, 
                    bottom=0.1, 
                    right=0.9, 
                    top=0.9, 
                    wspace=0.2, 
                    hspace=0.5)
plt.savefig("assets/graphiques/PREDICTION_1_Synthese.jpg", bbox_inches='tight')
plt.show()


plt.figure(figsize=(12,8))
for i in liste_pays_retenus:
    sns.lineplot(
        x=df_prediction[df_prediction['Indicator Code'] == 'IT.NET.USER.P2'][df_prediction['Country Name'] == i ]['Year'],
        y=df_prediction[df_prediction['Indicator Code'] == 'IT.NET.USER.P2'][df_prediction['Country Name'] == i ]['Value'],
        label = i,
        linewidth=2.5,
        legend=False,
    )

plt.legend(bbox_to_anchor=(1, 1), loc = 2)
plt.title("IT.NET.USER.P2", size=20,weight='bold')
plt.ylabel("%",rotation=0)
plt.xlabel("Années")
plt.grid(False)
plt.savefig("assets/graphiques/PREDICTION_2_IT.NET.USER.P2.jpg", bbox_inches='tight')
plt.show()


plt.figure(figsize=(12,8))
for i in liste_pays_retenus:
    sns.lineplot(
        x=df_prediction[df_prediction['Indicator Code'] == 'NY.GNP.PCAP.PP.CD'][df_prediction['Country Name'] == i ]['Year'],
        y=df_prediction[df_prediction['Indicator Code'] == 'NY.GNP.PCAP.PP.CD'][df_prediction['Country Name'] == i ]['Value'],
        label = i,
        linewidth=2.5,
        legend=False
    )
plt.legend(bbox_to_anchor=(1, 1), loc = 2)
plt.title("NY.GNP.PCAP.PP.CD", size=20,weight='bold')
plt.ylabel("",rotation=0)
plt.xlabel("Années")
plt.grid(False)
plt.savefig("assets/graphiques/PREDICTION_3_IT.NET.USER.P2.jpg", bbox_inches='tight')
plt.show()


plt.figure(figsize=(12,8))
for i in liste_pays_retenus:
    sns.lineplot(
        x=df_prediction[df_prediction['Indicator Code'] == 'SP.POP.1524.TO.UN'][df_prediction['Country Name'] == i ]['Year'],
        y=df_prediction[df_prediction['Indicator Code'] == 'SP.POP.1524.TO.UN'][df_prediction['Country Name'] == i ]['Value'],
        label = i,
        linewidth=2.5,
        legend=False
    )
plt.legend(bbox_to_anchor=(1, 1), loc = 2)
plt.title('SP.POP.1524.TO.UN', size=20,weight='bold')
plt.ylabel("",rotation=0)
plt.xlabel("Années")
plt.grid(False)
plt.savefig("assets/graphiques/PREDICTION_4_SP.POP.1524.TO.UN.jpg", bbox_inches='tight')
plt.show()


plt.figure(figsize=(12,8))
for i in liste_pays_retenus:
    sns.lineplot(
        x=df_prediction[df_prediction['Indicator Code'] == 'SE.SEC.ENRR'][df_prediction['Country Name'] == i ]['Year'],
        y=df_prediction[df_prediction['Indicator Code'] == 'SE.SEC.ENRR'][df_prediction['Country Name'] == i ]['Value'],
        label = i,
        linewidth=2.5,
        legend=False
    )
    
plt.legend(bbox_to_anchor=(1, 1), loc = 2)
plt.title("SE.SEC.ENRR", size=20,weight='bold')
plt.ylabel("%",rotation=0)
plt.xlabel("Années")
plt.grid(False)
plt.savefig("assets/graphiques/PREDICTION_5_SE.SEC.ENRR.jpg", bbox_inches='tight')
plt.show()


plt.figure(figsize=(12,8))
for i in liste_pays_retenus:
    sns.lineplot(
        x=df_prediction[df_prediction['Indicator Code'] == 'SE.TER.ENRR'][df_prediction['Country Name'] == i ]['Year'],
        y=df_prediction[df_prediction['Indicator Code'] == 'SE.TER.ENRR'][df_prediction['Country Name'] == i ]['Value'],
        label = i,
        linewidth=2.5
    )
plt.legend(bbox_to_anchor=(1, 1), loc = 2)
plt.title("SE.TER.ENRR", size=20,weight='bold')
plt.ylabel("%",rotation=0)
plt.xlabel("Années")
plt.grid(False)
plt.savefig("assets/graphiques/PREDICTION_6_SE.TER.ENRR.jpg", bbox_inches='tight')
plt.show()

	National accounts reference year	Latest industrial data	Latest trade data	Unnamed: 31
count	32.00000	107.000000	185.000000	0.0
mean	2001.53125	2008.102804	2010.994595	NaN
std	5.24856	2.616834	2.569675	NaN
min	1987.00000	2000.000000	1995.000000	NaN
25%	1996.75000	2007.500000	2011.000000	NaN
50%	2002.00000	2009.000000	2012.000000	NaN
75%	2005.00000	2010.000000	2012.000000	NaN
max	2012.00000	2010.000000	2012.000000	NaN

	CountryCode	SeriesCode	DESCRIPTION	Unnamed: 4
Year
1970	5374	5374	5374	0
1971	8507	8507	8507	0
1972	8661	8661	8661	0
1973	8571	8571	8571	0
1974	8512	8512	8512	0
1975	8627	8627	8627	0
1976	8729	8729	8729	0
1977	8593	8593	8593	0
1978	8585	8585	8585	0
1979	8742	8742	8742	0
1980	8950	8950	8950	0
1981	9465	9465	9465	0
1982	9014	9014	9014	0
1983	9227	9227	9227	0
1984	9386	9386	9386	0
1985	9653	9653	9653	0
1986	9743	9743	9743	0
1987	9357	9357	9357	0
1988	9382	9382	9382	0
1989	9517	9517	9517	0
1990	10616	10616	10616	0
1991	11129	11129	11129	0
1992	11309	11309	11309	0
1993	11093	11093	11093	0
1994	11448	11448	11448	0
1995	11194	11194	11194	0
1996	11197	11197	11197	0
1997	11167	11167	11167	0
1998	14916	14916	14916	0
1999	24938	24938	24938	0
2000	25093	25093	25093	0
2001	25047	25047	25047	0
2002	25687	25687	25687	0
2003	25683	25683	25683	0
2004	27128	27128	27128	0
2005	25992	25992	25992	0
2006	23394	23394	23394	0
2007	24263	24263	24263	0
2008	25018	25018	25018	0
2009	22965	22965	22965	0
2010	22598	22598	22598	0
2011	17760	17760	17760	0
2012	13904	13904	13904	0
2013	13405	13405	13405	0
2014	11378	11378	11378	0
2015	6348	6348	6348	0
2016	819	819	819	0
2017	558	558	558	0
2020	140	140	140	0
2025	140	140	140	0
2030	140	140	140	0
2035	144	144	144	0
2040	144	144	144	0
2045	144	144	144	0
2050	144	144	144	0

	1970	1971	1972	1973	1974	1975	1976	1977	1978	1979	...	2060	2065	2070	2075	2080	2085	2090	2095	2100	Unnamed: 69
count	7.228800e+04	3.553700e+04	3.561900e+04	3.554500e+04	3.573000e+04	8.730600e+04	3.748300e+04	3.757400e+04	3.757600e+04	3.680900e+04	...	5.143600e+04	5.143600e+04	5.143600e+04	5.143600e+04	5.143600e+04	5.143600e+04	5.143600e+04	5.143600e+04	5.143600e+04	0.0
mean	1.974772e+09	4.253638e+09	4.592365e+09	5.105006e+09	5.401493e+09	2.314288e+09	5.731808e+09	6.124437e+09	6.671489e+09	7.436724e+09	...	7.224868e+02	7.271290e+02	7.283779e+02	7.266484e+02	7.228327e+02	7.176899e+02	7.113072e+02	7.034274e+02	6.940296e+02	NaN
std	1.211687e+11	1.804814e+11	1.914083e+11	2.059170e+11	2.112150e+11	1.375059e+11	2.215546e+11	2.325489e+11	2.473986e+11	2.660957e+11	...	2.215845e+04	2.287990e+04	2.352338e+04	2.408149e+04	2.455897e+04	2.496587e+04	2.530183e+04	2.556069e+04	2.574189e+04	NaN
min	-1.435564e+00	-1.594625e+00	-3.056522e+00	-4.032582e+00	-4.213563e+00	-3.658569e+00	-2.950945e+00	-3.174870e+00	-3.558749e+00	-2.973612e+00	...	-1.630000e+00	-1.440000e+00	-1.260000e+00	-1.090000e+00	-9.200000e-01	-7.800000e-01	-6.500000e-01	-5.500000e-01	-4.500000e-01	NaN
25%	8.900000e-01	8.853210e+00	9.240920e+00	9.595200e+00	9.861595e+00	1.400000e+00	9.312615e+00	9.519913e+00	1.000000e+01	1.000000e+01	...	3.000000e-02	3.000000e-02	2.000000e-02	2.000000e-02	1.000000e-02	1.000000e-02	1.000000e-02	1.000000e-02	1.000000e-02	NaN
50%	6.317724e+00	6.316240e+01	6.655139e+01	6.969595e+01	7.087760e+01	9.677420e+00	7.101590e+01	7.133326e+01	7.290512e+01	7.510173e+01	...	2.300000e-01	2.300000e-01	2.300000e-01	2.300000e-01	2.300000e-01	2.300000e-01	2.300000e-01	2.300000e-01	2.200000e-01	NaN
75%	6.251250e+01	5.655200e+04	5.863650e+04	6.202900e+04	6.383675e+04	7.854163e+01	5.682800e+04	5.739175e+04	5.940425e+04	6.411500e+04	...	7.505000e+00	7.500000e+00	7.300000e+00	7.100000e+00	6.722500e+00	6.080000e+00	5.462500e+00	4.680000e+00	4.032500e+00	NaN
max	1.903929e+13	1.986457e+13	2.100916e+13	2.238367e+13	2.282991e+13	2.300634e+13	2.424128e+13	2.521383e+13	2.622101e+13	2.730873e+13	...	2.951569e+06	3.070879e+06	3.169711e+06	3.246239e+06	3.301586e+06	3.337871e+06	3.354746e+06	3.351887e+06	3.330484e+06	NaN

	nb_nonnull	%_nonnull	%_NaN
annees
2010	242442	27.33	72.67
2011	146012	16.46	83.54
2012	147264	16.60	83.40
2013	137509	15.50	84.50
2014	113789	12.83	87.17
2015	131058	14.78	85.22
2016	16460	1.86	98.14
2017	143	0.02	99.98

	count	mean	std	min	25%	50%	75%	max
nb_indicateur	3665.0	237.404366	302.391383	0.0	15.000000	135.000000	332.000000	1262.000000
%	3665.0	16.350163	20.825853	0.0	1.033058	9.297521	22.865014	86.914601

	Country Code	Short Name	Table Name	Long Name	2-alpha code	Currency Unit	Special Notes	Region	Income Group	WB-2 code	...	IMF data dissemination standard	Latest population census	Latest household survey	Source of most recent Income and expenditure data	Vital registration complete	Latest agricultural census	Latest industrial data	Latest trade data	Latest water withdrawal data	Unnamed: 31
0	ABW	Aruba	Aruba	Aruba	AW	Aruban florin	SNA data for 2000-2011 are updated from offici...	Latin America & Caribbean	High income: nonOECD	AW	...	NaN	2010	NaN	NaN	Yes	NaN	NaN	2012.0	NaN	NaN
1	AFG	Afghanistan	Afghanistan	Islamic State of Afghanistan	AF	Afghan afghani	Fiscal year end: March 20; reporting period fo...	South Asia	Low income	AF	...	General Data Dissemination System (GDDS)	1979	Multiple Indicator Cluster Survey (MICS), 2010/11	Integrated household survey (IHS), 2008	NaN	2013/14	NaN	2012.0	2000	NaN
2	AGO	Angola	Angola	People's Republic of Angola	AO	Angolan kwanza	April 2013 database update: Based on IMF data,...	Sub-Saharan Africa	Upper middle income	AO	...	General Data Dissemination System (GDDS)	1970	Malaria Indicator Survey (MIS), 2011	Integrated household survey (IHS), 2008	NaN	2015	NaN	NaN	2005	NaN
3	ALB	Albania	Albania	Republic of Albania	AL	Albanian lek	NaN	Europe & Central Asia	Upper middle income	AL	...	General Data Dissemination System (GDDS)	2011	Demographic and Health Survey (DHS), 2008/09	Living Standards Measurement Study Survey (LSM...	Yes	2012	2010.0	2012.0	2006	NaN
4	AND	Andorra	Andorra	Principality of Andorra	AD	Euro	NaN	Europe & Central Asia	High income: nonOECD	AD	...	NaN	2011. Population figures compiled from adminis...	NaN	NaN	Yes	NaN	NaN	2006.0	NaN	NaN
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
236	XKX	Kosovo	Kosovo	Republic of Kosovo	NaN	Euro	Kosovo became a World Bank member on June 29, ...	Europe & Central Asia	Lower middle income	KV	...	General Data Dissemination System (GDDS)	2011	NaN	Integrated household survey (IHS), 2011	NaN	NaN	NaN	NaN	NaN	NaN
237	YEM	Yemen	Yemen, Rep.	Republic of Yemen	YE	Yemeni rial	Based on official government statistics and In...	Middle East & North Africa	Lower middle income	RY	...	General Data Dissemination System (GDDS)	2004	Demographic and Health Survey (DHS), 2013	Expenditure survey/budget survey (ES/BS), 2005	NaN	NaN	2006.0	2012.0	2005	NaN
238	ZAF	South Africa	South Africa	Republic of South Africa	ZA	South African rand	Fiscal year end: March 31; reporting period fo...	Sub-Saharan Africa	Upper middle income	ZA	...	Special Data Dissemination Standard (SDDS)	2011	Demographic and Health Survey (DHS), 2003; Wor...	Expenditure survey/budget survey (ES/BS), 2010	NaN	2007	2010.0	2012.0	2000	NaN
239	ZMB	Zambia	Zambia	Republic of Zambia	ZM	New Zambian kwacha	National accounts data have rebased to reflect...	Sub-Saharan Africa	Lower middle income	ZM	...	General Data Dissemination System (GDDS)	2010	Demographic and Health Survey (DHS), 2013	Integrated household survey (IHS), 2010	NaN	2010. Population and Housing Census.	NaN	2011.0	2002	NaN
240	ZWE	Zimbabwe	Zimbabwe	Republic of Zimbabwe	ZW	U.S. dollar	Fiscal year end: June 30; reporting period for...	Sub-Saharan Africa	Low income	ZW	...	General Data Dissemination System (GDDS)	2012	Demographic and Health Survey (DHS), 2010/11	Integrated household survey (IHS), 2011/12	NaN	NaN	NaN	2012.0	2002	NaN

	Country Code	Short Name	Table Name	Long Name	2-alpha code	Currency Unit	Special Notes	Region	Income Group	WB-2 code	...	External debt Reporting status	System of trade	Government Accounting concept	IMF data dissemination standard	Latest population census	Latest household survey	Source of most recent Income and expenditure data	Vital registration complete	Latest agricultural census	Latest water withdrawal data
count	241	241	241	241	238	215	145	214	214	240	...	124	200	161	181	213	141	160	111	142	179
unique	241	241	241	241	238	152	131	7	5	240	...	3	2	2	2	27	60	75	2	35	20
top	ABW	Aruba	Aruba	Aruba	AW	Euro	April 2012 database update: Based on official ...	Europe & Central Asia	Upper middle income	AW	...	Actual	General trade system	Consolidated central government	General Data Dissemination System (GDDS)	2011	World Health Survey (WHS), 2003	Integrated household survey (IHS), 2012	Yes	2010	2000
freq	1	1	1	1	1	23	6	57	55	1	...	107	106	95	110	59	10	15	110	36	40

	Country Code	Short Name	Table Name	Long Name	2-alpha code	Currency Unit	Special Notes	Region	Income Group	WB-2 code	...	IMF data dissemination standard	Latest population census	Latest household survey	Source of most recent Income and expenditure data	Vital registration complete	Latest agricultural census	Latest industrial data	Latest trade data	Latest water withdrawal data	Unnamed: 31
5	ARB	Arab World	Arab World	Arab World	1A	NaN	Arab World aggregate. Arab World is composed o...	NaN	NaN	1A	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
57	EAP	East Asia & Pacific (developing only)	East Asia & Pacific	East Asia & Pacific (developing only)	4E	NaN	East Asia and Pacific regional aggregate (does...	NaN	NaN	4E	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
58	EAS	East Asia & Pacific (all income levels)	East Asia & Pacific (all income levels)	East Asia & Pacific (all income levels)	Z4	NaN	East Asia and Pacific regional aggregate (incl...	NaN	NaN	Z4	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
59	ECA	Europe & Central Asia (developing only)	Europe & Central Asia	Europe & Central Asia (developing only)	7E	NaN	Europe and Central Asia regional aggregate (do...	NaN	NaN	7E	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
60	ECS	Europe & Central Asia (all income levels)	Europe & Central Asia (all income levels)	Europe & Central Asia (all income levels)	Z7	NaN	Europe and Central Asia regional aggregate (in...	NaN	NaN	Z7	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
63	EMU	Euro area	Euro area	Euro area	XC	NaN	Euro area aggregate.	NaN	NaN	XC	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
68	EUU	European Union	European Union	European Union	EU	NaN	European Union aggregate.	NaN	NaN	EU	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
78	GIB	Gibraltar	Gibraltar	Gibraltar	GI	Gibraltar pound	NaN	NaN	NaN	GI	...	NaN	2012	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
89	HIC	High income	High income	High income	XD	NaN	High income group aggregate. High-income econo...	NaN	NaN	XD	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
92	HPC	Heavily indebted poor countries (HIPC)	Heavily indebted poor countries (HIPC)	Heavily indebted poor countries (HIPC)	XE	NaN	Heavily indebted poor countries aggregate.	NaN	NaN	XE	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
116	LAC	Latin America & Caribbean (developing only)	Latin America & Caribbean	Latin America & Caribbean (developing only)	XJ	NaN	Latin America and Caribbean regional aggregate...	NaN	NaN	XJ	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
122	LCN	Latin America & Caribbean (all income levels)	Latin America & Caribbean (all income levels)	Latin America & Caribbean (all income levels)	ZJ	NaN	Latin America and Caribbean regional aggregate...	NaN	NaN	ZJ	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
123	LDC	Least developed countries: UN classification	Least developed countries: UN classification	Least developed countries: UN classification	XL	NaN	Least developed countries (UN classification) ...	NaN	NaN	XL	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
124	LIC	Low income	Low income	Low income	XM	NaN	Low income group aggregate. Low-income economi...	NaN	NaN	XM	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
127	LMC	Lower middle income	Lower middle income	Lower middle income	XN	NaN	Lower middle income group aggregate. Lower-mid...	NaN	NaN	XN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
128	LMY	Low & middle income	Low & middle income	Low & middle income	XO	NaN	Low and middle income group aggregate (all dev...	NaN	NaN	XO	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
140	MEA	Middle East & North Africa (all income levels)	Middle East & North Africa (all income levels)	Middle East & North Africa (all income levels)	ZQ	NaN	Middle East and North Africa regional aggregat...	NaN	NaN	ZQ	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
143	MIC	Middle income	Middle income	Middle income	XP	NaN	Middle income group aggregate. Middle-income e...	NaN	NaN	XP	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
148	MNA	Middle East & North Africa (developing only)	Middle East & North Africa	Middle East & North Africa (developing only)	XQ	NaN	Middle East and North Africa regional aggregat...	NaN	NaN	XQ	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
157	NAC	North America	North America	North America	XU	NaN	North America regional aggregate. There are no...	NaN	NaN	XU	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
166	NRU	Nauru	Nauru	Nauru	NR	NaN	NaN	NaN	NaN	NR	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
168	OED	OECD members	OECD members	OECD members	OE	NaN	OECD members aggregate (all, including develop...	NaN	NaN	OE	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
187	SAS	South Asia	South Asia	South Asia	8S	NaN	South Asia regional aggregate. There are no ec...	NaN	NaN	8S	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
198	SSA	Sub-Saharan Africa (developing only)	Sub-Saharan Africa	Sub-Saharan Africa (developing only)	ZF	NaN	Sub-Saharan Africa regional aggregate (does no...	NaN	NaN	ZF	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
200	SSF	Sub-Saharan Africa (all income levels)	Sub-Saharan Africa (all income levels)	Sub-Saharan Africa (all income levels)	ZG	NaN	Sub-Saharan Africa regional aggregate (includi...	NaN	NaN	ZG	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
225	UMC	Upper middle income	Upper middle income	Upper middle income	XT	NaN	Upper middle income group aggregate. Upper-mid...	NaN	NaN	XT	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
234	WLD	World	World	World	1W	NaN	World aggregate.	NaN	NaN	1W	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN

	Currency Unit	Short Name
46	Euro	23
143	U.S. dollar	14
23	CFA franc	14
42	East Caribbean dollar	6
114	Pound sterling	3
38	Danish krone	3

	CountryCode	SeriesCode	DESCRIPTION	Unnamed: 3
0	ABW	SP.POP.TOTL	Data sources : United Nations World Population...	NaN
1	ABW	SP.POP.GROW	Data sources: United Nations World Population ...	NaN
2	AFG	SP.POP.GROW	Data sources: United Nations World Population ...	NaN
3	AFG	NY.GDP.PCAP.PP.CD	Estimates are based on regression.	NaN
4	AFG	SP.POP.TOTL	Data sources : United Nations World Population...	NaN

	Short Name
5	Arab World
47	Curaçao
57	East Asia & Pacific (developing only)
58	East Asia & Pacific (all income levels)
59	Europe & Central Asia (developing only)
60	Europe & Central Asia (all income levels)
63	Euro area
68	European Union
89	High income
92	Heavily indebted poor countries (HIPC)
116	Latin America & Caribbean (developing only)
122	Latin America & Caribbean (all income levels)
123	Least developed countries: UN classification
124	Low income
127	Lower middle income
128	Low & middle income
134	St. Martin (French part)
140	Middle East & North Africa (all income levels)
143	Middle income
148	Middle East & North Africa (developing only)
157	North America
166	Nauru
168	OECD members
187	South Asia
198	Sub-Saharan Africa (developing only)
199	South Sudan
200	Sub-Saharan Africa (all income levels)
207	Sint Maarten (Dutch part)
225	Upper middle income
234	World

	CountryCode	SeriesCode	Year	DESCRIPTION	Unnamed: 4
0	ABW	SE.PRE.ENRL.FE	YR2001	Country estimation.	NaN
1	ABW	SE.TER.TCHR.FE	YR2005	Country estimation.	NaN
2	ABW	SE.PRE.TCHR.FE	YR2000	Country estimation.	NaN
3	ABW	SE.SEC.ENRL.GC	YR2004	Country estimation.	NaN
4	ABW	SE.PRE.TCHR	YR2006	Country estimation.	NaN

	Series Code	Topic	Indicator Name	Short definition	Long definition	Unit of measure	Periodicity	Base Period	Other notes	Aggregation method	...	Notes from original source	General comments	Source	Statistical concept and methodology	Development relevance	Related source links	Other web links	Related indicators	License Type	Unnamed: 20
0	BAR.NOED.1519.FE.ZS	Attainment	Barro-Lee: Percentage of female population age...	Percentage of female population age 15-19 with...	Percentage of female population age 15-19 with...	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	Robert J. Barro and Jong-Wha Lee: http://www.b...	NaN	NaN	NaN	NaN	NaN	NaN	NaN
1	BAR.NOED.1519.ZS	Attainment	Barro-Lee: Percentage of population age 15-19 ...	Percentage of population age 15-19 with no edu...	Percentage of population age 15-19 with no edu...	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	Robert J. Barro and Jong-Wha Lee: http://www.b...	NaN	NaN	NaN	NaN	NaN	NaN	NaN
2	BAR.NOED.15UP.FE.ZS	Attainment	Barro-Lee: Percentage of female population age...	Percentage of female population age 15+ with n...	Percentage of female population age 15+ with n...	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	Robert J. Barro and Jong-Wha Lee: http://www.b...	NaN	NaN	NaN	NaN	NaN	NaN	NaN
3	BAR.NOED.15UP.ZS	Attainment	Barro-Lee: Percentage of population age 15+ wi...	Percentage of population age 15+ with no educa...	Percentage of population age 15+ with no educa...	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	Robert J. Barro and Jong-Wha Lee: http://www.b...	NaN	NaN	NaN	NaN	NaN	NaN	NaN
4	BAR.NOED.2024.FE.ZS	Attainment	Barro-Lee: Percentage of female population age...	Percentage of female population age 20-24 with...	Percentage of female population age 20-24 with...	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	Robert J. Barro and Jong-Wha Lee: http://www.b...	NaN	NaN	NaN	NaN	NaN	NaN	NaN

	Topic	Series Code
19	Learning Outcomes	1046
0	Attainment	733
9	Education Equality	426
28	Secondary	256
24	Primary	248
21	Population	213
34	Tertiary	158
32	Teachers	137
12	Expenditures	93
11	Engaging the Private Sector (SABER)	51
20	Literacy	32
33	Teachers (SABER)	29
25	School Autonomy and Accountability (SABER)	29
4	Early Childhood Education	24
10	Education Management Information Systems (SABER)	23
22	Post-Secondary/Non-Tertiary	19
26	School Finance (SABER)	18
35	Tertiary Education (SABER)	17
31	Student Assessment (SABER)	15
23	Pre-Primary	14
15	Health: Population: Structure	13
36	Workforce Development (SABER)	12
3	Early Child Development (SABER)	12
29	Social Protection & Labor: Labor force structure	11
27	School Health and School Feeding (SABER)	9
8	Economic Policy & Debt: Purchasing power parity	6
30	Social Protection & Labor: Unemployment	3
7	Economic Policy & Debt: National accounts: US$...	3
18	Laber	3
1	Background	2
17	Infrastructure: Communications	2
6	Economic Policy & Debt: National accounts: US$...	2
2	EMIS	2
16	Health: Risk factors	1
14	Health: Population: Dynamics	1
13	Health: Mortality	1
5	Economic Policy & Debt: National accounts: Atl...	1

	Country Name	Country Code	Indicator Name	Indicator Code	1970	1971	1972	1973	1974	1975	...	2060	2065	2070	2075	2080	2085	2090	2095	2100	Unnamed: 69
0	Arab World	ARB	Adjusted net enrolment rate, lower secondary, ...	UIS.NERA.2	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
1	Arab World	ARB	Adjusted net enrolment rate, lower secondary, ...	UIS.NERA.2.F	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
2	Arab World	ARB	Adjusted net enrolment rate, lower secondary, ...	UIS.NERA.2.GPI	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
3	Arab World	ARB	Adjusted net enrolment rate, lower secondary, ...	UIS.NERA.2.M	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
4	Arab World	ARB	Adjusted net enrolment rate, primary, both sex...	SE.PRM.TENR	54.822121	54.894138	56.209438	57.267109	57.991138	59.36554	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN

	2010	2011	2012	2013	2014	2015	nb_indicateur	%	Indicator Code
Indicator Code
IT.NET.USER.P2	202	204	202	201	201	201	1211	83.402204	IT.NET.USER.P2
NY.GDP.MKTP.CD	202	202	198	199	197	196	1194	82.231405	NY.GDP.MKTP.CD
NY.GDP.MKTP.KD	202	198	196	196	194	193	1179	81.198347	NY.GDP.MKTP.KD
NY.GDP.PCAP.CD	202	202	198	199	197	196	1194	82.231405	NY.GDP.PCAP.CD
NY.GDP.PCAP.KD	202	198	196	196	194	193	1179	81.198347	NY.GDP.PCAP.KD
SE.PRM.AGES	203	203	203	203	203	203	1218	83.884298	SE.PRM.AGES
SE.PRM.DURS	203	203	203	203	203	203	1218	83.884298	SE.PRM.DURS
SE.SEC.AGES	200	199	199	200	200	200	1198	82.506887	SE.SEC.AGES
SE.SEC.DURS	200	199	199	200	200	200	1198	82.506887	SE.SEC.DURS
SE.SEC.DURS.LO	200	199	199	200	200	200	1198	82.506887	SE.SEC.DURS.LO
SE.SEC.DURS.UP	200	199	199	200	200	200	1198	82.506887	SE.SEC.DURS.UP
SP.POP.GROW	211	210	210	210	210	210	1261	86.845730	SP.POP.GROW
SP.POP.TOTL	211	211	210	210	210	210	1262	86.914601	SP.POP.TOTL
UIS.THAGE.0	200	200	200	200	198	198	1196	82.369146	UIS.THAGE.0
UIS.THAGE.3.A.GPV	200	199	199	200	199	198	1195	82.300275	UIS.THAGE.3.A.GPV
UIS.THDUR.0	200	200	200	200	200	200	1200	82.644628	UIS.THDUR.0

	Series Code	Indicator Name	Topic	Long definition
2209	SE.ADT.1524.LT.FE.ZS	Youth literacy rate, population 15-24 years, female (%)	Literacy	Number of females age 15 to 24 years who can both read and write with understanding a short simple statement on their everyday life, divided by the female population in that age group. Generally, ‘literacy’ also encompasses ‘numeracy’, the ability to make simple arithmetic calculations. Divide the number of people aged 15 to 24 years who are literate by the total population in the same age group and multiply the result by 100.
2210	SE.ADT.1524.LT.FM.ZS	Youth literacy rate, population 15-24 years, gender parity index (GPI)	Literacy	Ratio of female youth literacy rate to male youth literacy rate. It is calculated by dividing the female value for the indicator by the male value for the indicator. A GPI equal to 1 indicates parity between females and males. In general, a value less than 1 indicates disparity in favor of males and a value greater than 1 indicates disparity in favor of females.
2211	SE.ADT.1524.LT.MA.ZS	Youth literacy rate, population 15-24 years, male (%)	Literacy	Number of males age 15 to 24 years who can both read and write with understanding a short simple statement on their everyday life, divided by the male population in that age group. Generally, ‘literacy’ also encompasses ‘numeracy’, the ability to make simple arithmetic calculations. Divide the number of people aged 15 to 24 years who are literate by the total population in the same age group and multiply the result by 100.
2212	SE.ADT.1524.LT.ZS	Youth literacy rate, population 15-24 years, both sexes (%)	Literacy	Number of people age 15 to 24 years who can both read and write with understanding a short simple statement on their everyday life, divided by the population in that age group. Generally, ‘literacy’ also encompasses ‘numeracy’, the ability to make simple arithmetic calculations. Divide the number of people aged 15 to 24 years who are literate by the total population in the same age group and multiply the result by 100.
2504	SP.POP.1524.FE.UN	Population, ages 15-24, female	Population	Population, ages 15-24, female is the total number of females age 15-24.
2505	SP.POP.1524.MA.UN	Population, ages 15-24, male	Population	Population, ages 15-24, male is the total number of males age 15-24.
2506	SP.POP.1524.TO.UN	Population, ages 15-24, total	Population	Population, ages 15-24, total is the total population age 15-24.

	Series Code	Indicator Name	Topic	Long definition
2294	SE.SEC.ENRL	Enrolment in secondary education, both sexes (number)	Secondary	Total number of students enrolled at public and private secondary education institutions regardless of age.
2295	SE.SEC.ENRL.FE	Enrolment in secondary education, female (number)	Secondary	Total number of female students enrolled at public and private secondary education institutions regardless of age.
2296	SE.SEC.ENRL.FE.VO.ZS	Percentage of female students in secondary education enrolled in vocational programmes, female (%)	Secondary	Total number of female students enrolled in vocational programmes at the secondary education level, expressed as a percentage of the total number of female students enrolled in all programmes (vocational and general) at the secondary level. Vocational education is designed for learners to acquire the knowledge, skills and competencies specific to a particular occupation or trade or class of occupations or trades. Vocational education may have work-based components (e.g. apprenticeships). Successful completion of such programmes leads to labour-market relevant vocational qualifications acknowledged as occupationally-oriented by the relevant national authorities and/or the labour market.
2297	SE.SEC.ENRL.FE.ZS	Percentage of students in secondary education who are female (%)	Secondary	Number of female students enrolled in all secondary education programmes expressed as a percentage of the total number of students (male and female) enrolled at the secondary education level in a given school year.
2298	SE.SEC.ENRL.GC	Enrolment in secondary general, both sexes (number)	Secondary	Total number of students enrolled in general programmes at public and private secondary education institutions regardless of age.
2299	SE.SEC.ENRL.GC.FE	Enrolment in secondary general, female (number)	Secondary	Total number of female students enrolled in general programmes at public and private secondary education institutions regardless of age.
2300	SE.SEC.ENRL.GC.FE.ZS	Percentage of students in secondary general education who are female (%)	Secondary	Number of female students enrolled in general programmes at the secondary education level expressed as a percentage of the total number of students (male and female) enrolled in general programmes at the secondary education level in a given school year.
2301	SE.SEC.ENRL.MA.VO.ZS	Percentage of male students in secondary education enrolled in vocational programmes, male (%)	Secondary	Total number of male students enrolled in vocational programmes at the secondary education level, expressed as a percentage of the total number of male students enrolled in all programmes (vocational and general) at the secondary level. Vocational education is designed for learners to acquire the knowledge, skills and competencies specific to a particular occupation or trade or class of occupations or trades. Vocational education may have work-based components (e.g. apprenticeships). Successful completion of such programmes leads to labour-market relevant vocational qualifications acknowledged as occupationally-oriented by the relevant national authorities and/or the labour market.
2302	SE.SEC.ENRL.TC.ZS	Pupil-teacher ratio in secondary education (headcount basis)	Teachers	Average number of pupils per teacher at a given level of education, based on headcounts of both pupils and teachers. Divide the total number of pupils enrolled at the specified level of education by the number of teachers at the same level. In computing and interpreting this indicator, one should take into account the existence of part-time teaching, school-shifts, multi-grade classes and other practices that may affect the precision and meaningfulness of pupil-teacher ratios. When feasible, the number of part-time teachers is converted to ‘full-time equivalent’ teachers; a double-shift teacher is counted twice, etc. Teachers are defined as persons whose professional activity involves the transmitting of knowledge, attitudes and skills that are stipulated in a formal curriculum programme to students enrolled in a formal educational institution.
2303	SE.SEC.ENRL.VO	Enrolment in secondary vocational, both sexes (number)	Secondary	Total number of students enrolled in vocational programmes at public and private secondary education institutions. Vocational education is designed for learners to acquire the knowledge, skills and competencies specific to a particular occupation or trade or class of occupations or trades. Vocational education may have work-based components (e.g. apprenticeships). Successful completion of such programmes leads to labour-market relevant vocational qualifications acknowledged as occupationally-oriented by the relevant national authorities and/or the labour market.
2304	SE.SEC.ENRL.VO.FE	Enrolment in secondary vocational, female (number)	Secondary	Total number of female students enrolled in vocational programmes at public and private secondary education institutions. Vocational education is designed for learners to acquire the knowledge, skills and competencies specific to a particular occupation or trade or class of occupations or trades. Vocational education may have work-based components (e.g. apprenticeships). Successful completion of such programmes leads to labour-market relevant vocational qualifications acknowledged as occupationally-oriented by the relevant national authorities and/or the labour market.
2305	SE.SEC.ENRL.VO.FE.ZS	Percentage of students in secondary vocational education who are female (%)	Secondary	Number of female students enrolled in vocational programmes at the secondary education level expressed as a percentage of the total number of students (male and female) enrolled in vocational programmes at the secondary education level in a given school year.
2306	SE.SEC.ENRL.VO.ZS	Percentage of students in secondary education enrolled in vocational programmes, both sexes (%)	Secondary	Total number of students enrolled in vocational programmes at the secondary education level, expressed as a percentage of the total number of students enrolled in all programmes (vocational and general) at the secondary level. Vocational education is designed for learners to acquire the knowledge, skills and competencies specific to a particular occupation or trade or class of occupations or trades. Vocational education may have work-based components (e.g. apprenticeships). Successful completion of such programmes leads to labour-market relevant vocational qualifications acknowledged as occupationally-oriented by the relevant national authorities and/or the labour market.

	Country Name	Country Code	Indicator Name	Indicator Code
count	886930	886930	886930	886930
unique	242	242	3665	3665
top	Arab World	ARB	Adjusted net enrolment rate, lower secondary, ...	UIS.NERA.2
freq	3665	3665	242	242

	Region	0
0	North America	11386
1	South Asia	36446
2	Middle East & North Africa	95397
3	East Asia & Pacific	118810
4	Latin America & Caribbean	157163
5	Sub-Saharan Africa	205030
6	Europe & Central Asia	263389

	2010	2011	2012	2013	2014	2015	nb_indicateur	%
Indicator Code
BAR.NOED.1519.FE.ZS	144	0	0	0	0	0	144	9.917355
BAR.NOED.1519.ZS	144	0	0	0	0	0	144	9.917355
BAR.NOED.15UP.FE.ZS	144	0	0	0	0	0	144	9.917355
BAR.NOED.15UP.ZS	144	0	0	0	0	0	144	9.917355
BAR.NOED.2024.FE.ZS	144	0	0	0	0	0	144	9.917355

	Series Code	Indicator Name	Topic	Long definition
1666	NY.GNP.MKTP.CD	GNI (current US$)	Economic Policy & Debt: National accounts: US$ at current prices: Aggregate indicators	GNI (formerly GNP) is the sum of value added by all resident producers plus any product taxes (less subsidies) not included in the valuation of output plus net receipts of primary income (compensation of employees and property income) from abroad. Data are in current U.S. dollars.
1667	NY.GNP.MKTP.PP.CD	GNI, PPP (current international $)	Economic Policy & Debt: Purchasing power parity	PPP GNI (formerly PPP GNP) is gross national income (GNI) converted to international dollars using purchasing power parity rates. An international dollar has the same purchasing power over GNI as a U.S. dollar has in the United States. Gross national income is the sum of value added by all resident producers plus any product taxes (less subsidies) not included in the valuation of output plus net receipts of primary income (compensation of employees and property income) from abroad. Data are in current international dollars. For most economies PPP figures are extrapolated from the 2011 International Comparison Program (ICP) benchmark estimates or imputed using a statistical model based on the 2011 ICP. For 47 high- and upper middle-income economies conversion factors are provided by Eurostat and the Organisation for Economic Co-operation and Development (OECD).
1668	NY.GNP.PCAP.CD	GNI per capita, Atlas method (current US$)	Economic Policy & Debt: National accounts: Atlas GNI & GNI per capita	GNI per capita (formerly GNP per capita) is the gross national income, converted to U.S. dollars using the World Bank Atlas method, divided by the midyear population. GNI is the sum of value added by all resident producers plus any product taxes (less subsidies) not included in the valuation of output plus net receipts of primary income (compensation of employees and property income) from abroad. GNI, calculated in national currency, is usually converted to U.S. dollars at official exchange rates for comparisons across economies, although an alternative rate is used when the official exchange rate is judged to diverge by an exceptionally large margin from the rate actually applied in international transactions. To smooth fluctuations in prices and exchange rates, a special Atlas method of conversion is used by the World Bank. This applies a conversion factor that averages the exchange rate for a given year and the two preceding years, adjusted for differences in rates of inflation between the country, and through 2000, the G-5 countries (France, Germany, Japan, the United Kingdom, and the United States). From 2001, these countries include the Euro area, Japan, the United Kingdom, and the United States.
1669	NY.GNP.PCAP.PP.CD	GNI per capita, PPP (current international $)	Economic Policy & Debt: Purchasing power parity	GNI per capita based on purchasing power parity (PPP). PPP GNI is gross national income (GNI) converted to international dollars using purchasing power parity rates. An international dollar has the same purchasing power over GNI as a U.S. dollar has in the United States. GNI is the sum of value added by all resident producers plus any product taxes (less subsidies) not included in the valuation of output plus net receipts of primary income (compensation of employees and property income) from abroad. Data are in current international dollars based on the 2011 ICP round.

	Series Code	Indicator Name	Topic	Long definition
2332	SE.TER.ENRL	Enrolment in tertiary education, all programmes, both sexes (number)	Tertiary	The total number of students enrolled at public and private tertiary education institutions.
2333	SE.TER.ENRL.FE	Enrolment in tertiary education, all programmes, female (number)	Tertiary	The total number of female students enrolled at public and private tertiary education institutions.
2334	SE.TER.ENRL.FE.ZS	Percentage of students in tertiary education who are female (%)	Tertiary	Number of female students at the tertiary education level (ISCED 5 to 8) expressed as a percentage of the total number of students (male and female) at the tertiary education level (ISCED 5 to 8) in a given school year.
2335	SE.TER.ENRR	Gross enrolment ratio, tertiary, both sexes (%)	Tertiary	Total enrollment in tertiary education (ISCED 5 to 8), regardless of age, expressed as a percentage of the total population of the five-year age group following on from secondary school leaving.
2336	SE.TER.ENRR.FE	Gross enrolment ratio, tertiary, female (%)	Tertiary	Total female enrollment in tertiary education (ISCED 5 to 8), regardless of age, expressed as a percentage of the total female population of the five-year age group following on from secondary school leaving.
2337	SE.TER.ENRR.MA	Gross enrolment ratio, tertiary, male (%)	Tertiary	Total male enrollment in tertiary education (ISCED 5 to 8), regardless of age, expressed as a percentage of the total male population of the five-year age group following on from secondary school leaving.

	Indicator Name	Indicator Code	Country Name	Country Code	Region	Income Group	2010	2011	2012	2013	2014	2015
4	Population, ages 15-24, total	SP.POP.1524.TO.UN	211	211	210	210	181	181	181	181	181	181
3	Internet users (per 100 people)	IT.NET.USER.P2	211	211	210	210	202	204	202	201	201	201
2	Gross enrolment ratio, tertiary, both sexes (%)	SE.TER.ENRR	211	211	210	210	133	138	136	124	119	99
1	Gross enrolment ratio, secondary, both sexes (%)	SE.SEC.ENRR	211	211	210	210	145	148	142	133	133	116
0	GNI per capita, PPP (current international $)	NY.GNP.PCAP.PP.CD	211	211	210	210	187	187	185	185	183	181

	Country Name	Country Code	Indicator Name
103872	American Samoa	ASM	GNI per capita, PPP (current international $)
103955	American Samoa	ASM	Gross enrolment ratio, secondary, both sexes (%)
103959	American Samoa	ASM	Gross enrolment ratio, tertiary, both sexes (%)
103995	American Samoa	ASM	Internet users (per 100 people)
105102	American Samoa	ASM	Population, ages 15-24, total
107537	Andorra	AND	GNI per capita, PPP (current international $)
107620	Andorra	AND	Gross enrolment ratio, secondary, both sexes (%)
107624	Andorra	AND	Gross enrolment ratio, tertiary, both sexes (%)
108767	Andorra	AND	Population, ages 15-24, total
116097	Antigua and Barbuda	ATG	Population, ages 15-24, total
125862	Aruba	ABW	GNI per capita, PPP (current international $)
136940	Azerbaijan	AZE	Gross enrolment ratio, secondary, both sexes (%)
140609	Bahamas, The	BHS	Gross enrolment ratio, tertiary, both sexes (%)
171072	Bermuda	BMU	Population, ages 15-24, total
177259	Bolivia	BOL	Gross enrolment ratio, tertiary, both sexes (%)
180920	Bosnia and Herzegovina	BIH	Gross enrolment ratio, secondary, both sexes (%)
180924	Bosnia and Herzegovina	BIH	Gross enrolment ratio, tertiary, both sexes (%)
184585	Botswana	BWA	Gross enrolment ratio, secondary, both sexes (%)
213905	Cambodia	KHM	Gross enrolment ratio, secondary, both sexes (%)
221239	Canada	CAN	Gross enrolment ratio, tertiary, both sexes (%)
224817	Cayman Islands	CYM	GNI per capita, PPP (current international $)
224900	Cayman Islands	CYM	Gross enrolment ratio, secondary, both sexes (%)
224904	Cayman Islands	CYM	Gross enrolment ratio, tertiary, both sexes (%)
226047	Cayman Islands	CYM	Population, ages 15-24, total
235812	Channel Islands	CHI	GNI per capita, PPP (current international $)
235895	Channel Islands	CHI	Gross enrolment ratio, secondary, both sexes (%)
235899	Channel Islands	CHI	Gross enrolment ratio, tertiary, both sexes (%)
235935	Channel Islands	CHI	Internet users (per 100 people)
237042	Channel Islands	CHI	Population, ages 15-24, total
272462	Cuba	CUB	GNI per capita, PPP (current international $)
290787	Djibouti	DJI	GNI per capita, PPP (current international $)
294539	Dominica	DMA	Gross enrolment ratio, tertiary, both sexes (%)
295682	Dominica	DMA	Population, ages 15-24, total
312860	Equatorial Guinea	GNQ	Gross enrolment ratio, secondary, both sexes (%)
312864	Equatorial Guinea	GNQ	Gross enrolment ratio, tertiary, both sexes (%)
327437	Faroe Islands	FRO	GNI per capita, PPP (current international $)
327520	Faroe Islands	FRO	Gross enrolment ratio, secondary, both sexes (%)
327524	Faroe Islands	FRO	Gross enrolment ratio, tertiary, both sexes (%)
328667	Faroe Islands	FRO	Population, ages 15-24, total
331189	Fiji	FJI	Gross enrolment ratio, tertiary, both sexes (%)
342097	French Polynesia	PYF	GNI per capita, PPP (current international $)
342180	French Polynesia	PYF	Gross enrolment ratio, secondary, both sexes (%)
342184	French Polynesia	PYF	Gross enrolment ratio, tertiary, both sexes (%)
343327	French Polynesia	PYF	Population, ages 15-24, total
345845	Gabon	GAB	Gross enrolment ratio, secondary, both sexes (%)
345849	Gabon	GAB	Gross enrolment ratio, tertiary, both sexes (%)
364087	Gibraltar	GIB	GNI per capita, PPP (current international $)
364170	Gibraltar	GIB	Gross enrolment ratio, secondary, both sexes (%)
364174	Gibraltar	GIB	Gross enrolment ratio, tertiary, both sexes (%)
365317	Gibraltar	GIB	Population, ages 15-24, total
371417	Greenland	GRL	GNI per capita, PPP (current international $)
371500	Greenland	GRL	Gross enrolment ratio, secondary, both sexes (%)
371504	Greenland	GRL	Gross enrolment ratio, tertiary, both sexes (%)
372647	Greenland	GRL	Population, ages 15-24, total
378747	Guam	GUM	GNI per capita, PPP (current international $)
378830	Guam	GUM	Gross enrolment ratio, secondary, both sexes (%)
378834	Guam	GUM	Gross enrolment ratio, tertiary, both sexes (%)
379977	Guam	GUM	Population, ages 15-24, total
389825	Guinea-Bissau	GNB	Gross enrolment ratio, secondary, both sexes (%)
389829	Guinea-Bissau	GNB	Gross enrolment ratio, tertiary, both sexes (%)
397155	Haiti	HTI	Gross enrolment ratio, secondary, both sexes (%)
397159	Haiti	HTI	Gross enrolment ratio, tertiary, both sexes (%)
426475	Iraq	IRQ	Gross enrolment ratio, secondary, both sexes (%)
426479	Iraq	IRQ	Gross enrolment ratio, tertiary, both sexes (%)
433722	Isle of Man	IMN	GNI per capita, PPP (current international $)
433805	Isle of Man	IMN	Gross enrolment ratio, secondary, both sexes (%)
433809	Isle of Man	IMN	Gross enrolment ratio, tertiary, both sexes (%)
433845	Isle of Man	IMN	Internet users (per 100 people)
434952	Isle of Man	IMN	Population, ages 15-24, total
459460	Kenya	KEN	Gross enrolment ratio, secondary, both sexes (%)
459464	Kenya	KEN	Gross enrolment ratio, tertiary, both sexes (%)
463125	Kiribati	KIR	Gross enrolment ratio, secondary, both sexes (%)
463129	Kiribati	KIR	Gross enrolment ratio, tertiary, both sexes (%)
464272	Kiribati	KIR	Population, ages 15-24, total
466707	Korea, Dem. People’s Rep.	PRK	GNI per capita, PPP (current international $)
474120	Kosovo	XKX	Gross enrolment ratio, secondary, both sexes (%)
474124	Kosovo	XKX	Gross enrolment ratio, tertiary, both sexes (%)
474160	Kosovo	XKX	Internet users (per 100 people)
475267	Kosovo	XKX	Population, ages 15-24, total
503440	Libya	LBY	Gross enrolment ratio, secondary, both sexes (%)
503444	Libya	LBY	Gross enrolment ratio, tertiary, both sexes (%)
507022	Liechtenstein	LIE	GNI per capita, PPP (current international $)
508252	Liechtenstein	LIE	Population, ages 15-24, total
536425	Maldives	MDV	Gross enrolment ratio, secondary, both sexes (%)
548567	Marshall Islands	MHL	Population, ages 15-24, total
562080	Micronesia, Fed. Sts.	FSM	Gross enrolment ratio, secondary, both sexes (%)
562084	Micronesia, Fed. Sts.	FSM	Gross enrolment ratio, tertiary, both sexes (%)
569327	Monaco	MCO	GNI per capita, PPP (current international $)
569410	Monaco	MCO	Gross enrolment ratio, secondary, both sexes (%)
569414	Monaco	MCO	Gross enrolment ratio, tertiary, both sexes (%)
570557	Monaco	MCO	Population, ages 15-24, total
577887	Montenegro	MNE	Population, ages 15-24, total
591400	Namibia	NAM	Gross enrolment ratio, secondary, both sexes (%)
591404	Namibia	NAM	Gross enrolment ratio, tertiary, both sexes (%)
605977	New Caledonia	NCL	GNI per capita, PPP (current international $)
606060	New Caledonia	NCL	Gross enrolment ratio, secondary, both sexes (%)
606064	New Caledonia	NCL	Gross enrolment ratio, tertiary, both sexes (%)
607207	New Caledonia	NCL	Population, ages 15-24, total
613394	Nicaragua	NIC	Gross enrolment ratio, tertiary, both sexes (%)
624302	Northern Mariana Islands	MNP	GNI per capita, PPP (current international $)
624385	Northern Mariana Islands	MNP	Gross enrolment ratio, secondary, both sexes (%)
624389	Northern Mariana Islands	MNP	Gross enrolment ratio, tertiary, both sexes (%)
624425	Northern Mariana Islands	MNP	Internet users (per 100 people)
625532	Northern Mariana Islands	MNP	Population, ages 15-24, total
631719	Oman	OMN	Gross enrolment ratio, tertiary, both sexes (%)
639085	Palau	PLW	Internet users (per 100 people)
640192	Palau	PLW	Population, ages 15-24, total
646379	Papua New Guinea	PNG	Gross enrolment ratio, tertiary, both sexes (%)
669512	Puerto Rico	PRI	Population, ages 15-24, total
686694	Samoa	WSM	Gross enrolment ratio, tertiary, both sexes (%)
690272	San Marino	SMR	GNI per capita, PPP (current international $)
691502	San Marino	SMR	Population, ages 15-24, total
706162	Serbia	SRB	Population, ages 15-24, total
709827	Seychelles	SYC	Population, ages 15-24, total
712349	Sierra Leone	SLE	Gross enrolment ratio, tertiary, both sexes (%)
716010	Singapore	SGP	Gross enrolment ratio, secondary, both sexes (%)
716014	Singapore	SGP	Gross enrolment ratio, tertiary, both sexes (%)
730674	Solomon Islands	SLB	Gross enrolment ratio, tertiary, both sexes (%)
734252	Somalia	SOM	GNI per capita, PPP (current international $)
734335	Somalia	SOM	Gross enrolment ratio, secondary, both sexes (%)
734339	Somalia	SOM	Gross enrolment ratio, tertiary, both sexes (%)
753807	St. Kitts and Nevis	KNA	Population, ages 15-24, total
763659	St. Vincent and the Grenadines	VCT	Gross enrolment ratio, tertiary, both sexes (%)
770989	Suriname	SUR	Gross enrolment ratio, tertiary, both sexes (%)
785562	Syrian Arab Republic	SYR	GNI per capita, PPP (current international $)
807639	Tonga	TON	Gross enrolment ratio, tertiary, both sexes (%)
811300	Trinidad and Tobago	TTO	Gross enrolment ratio, secondary, both sexes (%)
811304	Trinidad and Tobago	TTO	Gross enrolment ratio, tertiary, both sexes (%)
825877	Turks and Caicos Islands	TCA	GNI per capita, PPP (current international $)
825960	Turks and Caicos Islands	TCA	Gross enrolment ratio, secondary, both sexes (%)
825964	Turks and Caicos Islands	TCA	Gross enrolment ratio, tertiary, both sexes (%)
826000	Turks and Caicos Islands	TCA	Internet users (per 100 people)
827107	Turks and Caicos Islands	TCA	Population, ages 15-24, total
829629	Tuvalu	TUV	Gross enrolment ratio, tertiary, both sexes (%)
830772	Tuvalu	TUV	Population, ages 15-24, total
840620	United Arab Emirates	ARE	Gross enrolment ratio, secondary, both sexes (%)
840624	United Arab Emirates	ARE	Gross enrolment ratio, tertiary, both sexes (%)
858949	Vanuatu	VUT	Gross enrolment ratio, tertiary, both sexes (%)
862614	Venezuela, RB	VEN	Gross enrolment ratio, tertiary, both sexes (%)
866275	Vietnam	VNM	Gross enrolment ratio, secondary, both sexes (%)
869857	Virgin Islands (U.S.)	VIR	GNI per capita, PPP (current international $)
869940	Virgin Islands (U.S.)	VIR	Gross enrolment ratio, secondary, both sexes (%)
869944	Virgin Islands (U.S.)	VIR	Gross enrolment ratio, tertiary, both sexes (%)
871087	Virgin Islands (U.S.)	VIR	Population, ages 15-24, total
873522	West Bank and Gaza	PSE	GNI per capita, PPP (current international $)
880935	Zambia	ZMB	Gross enrolment ratio, secondary, both sexes (%)

	Country Name	Country Code	Region	Indicator Name	Indicator Code	Income Group	2010	2011	2012	2013	2014	2015	Last_data
92877	Afghanistan	AFG	South Asia	GNI per capita, PPP (current international $)	NY.GNP.PCAP.PP.CD	Low income	1.590000e+03	1.680000e+03	1.900000e+03	1.900000e+03	1.890000e+03	1.880000e+03	1.880000e+03
92960	Afghanistan	AFG	South Asia	Gross enrolment ratio, secondary, both sexes (%)	SE.SEC.ENRR	Low income	5.324683e+01	5.461618e+01	5.667734e+01	5.668866e+01	5.565616e+01	5.564441e+01	5.564441e+01
92964	Afghanistan	AFG	South Asia	Gross enrolment ratio, tertiary, both sexes (%)	SE.TER.ENRR	Low income	NaN	3.755980e+00	NaN	NaN	8.662800e+00	NaN	NaN
93000	Afghanistan	AFG	South Asia	Internet users (per 100 people)	IT.NET.USER.P2	Low income	4.000000e+00	5.000000e+00	5.454545e+00	5.900000e+00	7.000000e+00	8.260000e+00	8.260000e+00
94107	Afghanistan	AFG	South Asia	Population, ages 15-24, total	SP.POP.1524.TO.UN	Low income	6.090239e+06	6.330684e+06	6.569239e+06	6.803785e+06	7.032072e+06	7.252785e+06	7.252785e+06

	Statistique	IT_last_data	ECO_last_data	SEC_last_data	TER_last_data	POP_last_data
0	mean	48.462239	1.998679e+04	83.988968	40.205176	8.038740e+06
1	median	50.219659	1.201000e+04	91.317413	36.794924	1.703510e+06
2	var	815.902838	4.648906e+08	893.296681	831.019710	7.317407e+14
3	std	28.564013	2.156132e+04	29.888069	28.827412	2.705071e+07
4	skew	-0.009762	1.885302e+00	-0.250500	0.367150	7.317494e+00
5	kurtosis	-1.305362	4.559860e+00	-0.340264	-1.114788	5.761620e+01

Indicator Code	IT.NET.USER.P2	NY.GNP.PCAP.PP.CD	SE.SEC.ENRR	SE.TER.ENRR	SP.POP.1524.TO.UN
Country Name
Afghanistan	8.260000	1880.0	55.644409	8.662800	7252785.0
Albania	63.252933	11340.0	95.765488	58.109951	556269.0
Algeria	38.200000	14170.0	99.860191	36.922279	6467818.0
Angola	12.400000	6270.0	28.898720	9.308020	4259352.0
Argentina	68.043064	19990.0	106.777901	82.917389	6886530.0

Introduction¶

PARTIE 1 : EDA - ANALYSE DES DONNES EXISTANTES¶

EDA - fonctions¶

Analyse du jeu de données EdStatsCountry¶

Description générale¶

Données manquantes¶

Les régions¶

Les pays et les régions¶

Income Group¶

Les monnaies utilisées¶

Visualiser les données Regions, Pays, Income Groups¶

Analyse du jeu de données EdStatsCountry-Series¶

Description générale¶

Données manquantes¶

CountryCode¶

SerieCode¶

DESCRIPTION¶

Visualisation Code Pays, Code Indicateur, Description¶

Analyse du jeu de données EdStatsFootNot¶

Description générale¶

Données Manquantes¶

CountryCode¶

DESCRIPTION¶

Year¶

Analyse du jeu de données EdStatsSeries¶

Descriptions générales¶

Données manquantes¶

Serie Code¶

Topic¶

Indicator Name¶

Long definition¶

Short definition¶

Les topics avec le plus d'indicateurs¶

Analyse du jeu de données EdStatsData¶

Decription générale¶

Données manquantes¶

Indicator Name, Indicator Name, Country, Code, Country Name¶

Comparer les pays presents dans les dataframes¶

Comparaison des pays présents dans EdStatsData.csv et EdStatsCountries.csv¶

Comparaison des pays présents dans EdStatsData.csv et EdStatsCountries-Series.csv¶

Données et années¶

Découpage en decennies¶

Repartition des données par région pour la décénnie 2010s¶

Decennie selectionnée 2010s¶

Taux de remplissage années 2010-2017¶

Bilan du dataset¶

PARTIE 2 : PREPROCESSING¶

Sélectionner les indicateurs pertinents¶

NA : Taux de remplissage par indicateur¶

Recherche des indicateurs pertinents par mots clés¶

NA par variable retenue¶

Indicateurs retenus :¶

Préparation données¶

Données disponible pour nos indicateurs¶

Gestion des NA¶

Identification et estimation des pays pour lesquels on a aucune donnée¶

Récupération des dernières données¶

On identifie de quelle année viennent les données¶

Boucle de collecte des dernières données par pays¶

Analyse des indicateurs retenus¶

Mesures de distribution¶

Mesures de tendance centrales¶

Identifier les outliers par indicateur¶

Gestion des indicateurs en % > 100%¶

Bilan analyse indicateurs retenus brut¶

Normaliser/Standardiser¶

Normaliser en supprimant les < median et suppression des SEC et TER à 100%¶

Déterminer des ordres de grandeurs des indicateurs statistiques classiques pour les différentes zones géographiques et pays du monde¶

Ordre de grandeur des indicateurs statistiques par pays¶

Déterminer des ordres de grandeurs des indicateurs statistiques classiques par region¶

PARTIE 3 : Scoring & MAP¶

Score d'attractivité pays¶

TOP 10 DF_SCORE¶

TOP 10 DF_POWERTRANSFORM_NORM¶

Score attractivité région¶

MAP¶

Evolution des indicateurs des pays du top 10¶

Indicator Code	IT.NET.USER.P2	NY.GNP.PCAP.PP.CD	SE.SEC.ENRR	SE.TER.ENRR	SP.POP.1524.TO.UN
count	140.000000	140.000000	140.000000	140.000000	1.400000e+02
mean	48.462239	19986.785714	83.988968	40.205176	8.038740e+06
std	28.666577	21638.740547	29.995388	28.930922	2.714784e+07
min	1.083733	670.000000	17.378010	0.797730	2.353600e+04
25%	21.845365	3977.500000	60.991915	12.270108	5.681078e+05
50%	50.219659	12010.000000	91.317413	36.794924	1.703510e+06
75%	72.949675	28492.500000	102.481300	64.801764	5.751938e+06
max	98.200000	125000.000000	166.808472	113.871788	2.441202e+08

	Country Name	Country Code	Region	Indicator Name	Indicator Code	Income Group	Last_data
85	Brunei Darussalam	BRN	East Asia & Pacific	GNI per capita, PPP (current international $)	NY.GNP.PCAP.PP.CD	High income: nonOECD	83600.0
350	Kuwait	KWT	Middle East & North Africa	GNI per capita, PPP (current international $)	NY.GNP.PCAP.PP.CD	High income: nonOECD	81150.0
390	Luxembourg	LUX	Europe & Central Asia	GNI per capita, PPP (current international $)	NY.GNP.PCAP.PP.CD	High income: OECD	68780.0
395	Macao SAR, China	MAC	East Asia & Pacific	GNI per capita, PPP (current international $)	NY.GNP.PCAP.PP.CD	High income: nonOECD	98650.0
495	Norway	NOR	Europe & Central Asia	GNI per capita, PPP (current international $)	NY.GNP.PCAP.PP.CD	High income: OECD	65180.0
535	Qatar	QAT	Middle East & North Africa	GNI per capita, PPP (current international $)	NY.GNP.PCAP.PP.CD	High income: nonOECD	125000.0
615	Switzerland	CHE	Europe & Central Asia	GNI per capita, PPP (current international $)	NY.GNP.PCAP.PP.CD	High income: OECD	63930.0

	Country Name	Country Code	Region	Indicator Name	Indicator Code	Income Group	Last_data
299	India	IND	South Asia	Population, ages 15-24, total	SP.POP.1524.TO.UN	Lower middle income	244120201.0
134	China	CHN	East Asia & Pacific	Population, ages 15-24, total	SP.POP.1524.TO.UN	Upper middle income	197026759.0
679	United States	USA	North America	Population, ages 15-24, total	SP.POP.1524.TO.UN	High income: OECD	45147517.0
304	Indonesia	IDN	East Asia & Pacific	Population, ages 15-24, total	SP.POP.1524.TO.UN	Lower middle income	41819264.0
504	Pakistan	PAK	South Asia	Population, ages 15-24, total	SP.POP.1524.TO.UN	Lower middle income	39080811.0
494	Nigeria	NGA	Sub-Saharan Africa	Population, ages 15-24, total	SP.POP.1524.TO.UN	Lower middle income	36659023.0
49	Bangladesh	BGD	South Asia	Population, ages 15-24, total	SP.POP.1524.TO.UN	Low income	34400594.0
84	Brazil	BRA	Latin America & Caribbean	Population, ages 15-24, total	SP.POP.1524.TO.UN	Upper middle income	33595574.0
219	Ethiopia	ETH	Sub-Saharan Africa	Population, ages 15-24, total	SP.POP.1524.TO.UN	Low income	21074230.0
444	Mexico	MEX	Latin America & Caribbean	Population, ages 15-24, total	SP.POP.1524.TO.UN	Upper middle income	20336734.0
524	Philippines	PHL	East Asia & Pacific	Population, ages 15-24, total	SP.POP.1524.TO.UN	Lower middle income	19278458.0
149	Congo, Dem. Rep.	COD	Sub-Saharan Africa	Population, ages 15-24, total	SP.POP.1524.TO.UN	Low income	15535095.0
199	Egypt, Arab Rep.	EGY	Middle East & North Africa	Population, ages 15-24, total	SP.POP.1524.TO.UN	Lower middle income	15377746.0
549	Russian Federation	RUS	Europe & Central Asia	Population, ages 15-24, total	SP.POP.1524.TO.UN	High income: nonOECD	14661984.0
654	Turkey	TUR	Europe & Central Asia	Population, ages 15-24, total	SP.POP.1524.TO.UN	Upper middle income	13930222.0
309	Iran, Islamic Rep.	IRN	Middle East & North Africa	Population, ages 15-24, total	SP.POP.1524.TO.UN	Upper middle income	13556577.0
334	Japan	JPN	East Asia & Pacific	Population, ages 15-24, total	SP.POP.1524.TO.UN	High income: OECD	12157426.0
584	South Africa	ZAF	Sub-Saharan Africa	Population, ages 15-24, total	SP.POP.1524.TO.UN	Upper middle income	9998431.0
629	Tanzania	TZA	Sub-Saharan Africa	Population, ages 15-24, total	SP.POP.1524.TO.UN	Low income	9849521.0
604	Sudan	SDN	Sub-Saharan Africa	Population, ages 15-24, total	SP.POP.1524.TO.UN	Lower middle income	9381489.0
634	Thailand	THA	East Asia & Pacific	Population, ages 15-24, total	SP.POP.1524.TO.UN	Upper middle income	9149718.0
139	Colombia	COL	Latin America & Caribbean	Population, ages 15-24, total	SP.POP.1524.TO.UN	Upper middle income	8999338.0
469	Myanmar	MMR	East Asia & Pacific	Population, ages 15-24, total	SP.POP.1524.TO.UN	Low income	8786817.0
244	Germany	DEU	Europe & Central Asia	Population, ages 15-24, total	SP.POP.1524.TO.UN	High income: OECD	8682394.0
664	Uganda	UGA	Sub-Saharan Africa	Population, ages 15-24, total	SP.POP.1524.TO.UN	Low income	8244246.0

Indicator	code	Pondérations
IT	IT.NET.USER.P2	5
POP 15-24	SP.POP.1524.TO.UN	4
SUPERIEUR	SE.TER.ENRR	3
SECONDAIRE	SE.SEC.ENRR	2
ECONOMIE	NY.GNP.PCAP.PP.CD	1

	IT.NET.USER.P2	NY.GNP.PCAP.PP.CD	SE.SEC.ENRR	SE.TER.ENRR	SP.POP.1524.TO.UN	Pays	Score
135	0.904686	1.462074	0.438005	1.345030	1.876738	United States	17.05
134	1.385123	1.126017	1.506951	0.690003	0.852785	United Kingdom	15.50
6	1.184173	1.222933	1.859802	1.433485	0.298286	Australia	15.22
69	1.322187	0.957368	0.483686	1.488486	0.749742	Korea, Rep.	15.11
66	1.359969	1.150131	0.583276	0.858930	1.112840	Japan	15.08
117	1.021567	0.963910	1.578976	1.421164	0.516456	Spain	14.66
12	1.197643	1.225723	2.938225	1.122505	-0.179942	Belgium	14.60
95	1.377759	1.310942	1.783850	1.196069	0.088065	Netherlands	14.49
109	0.871983	0.586230	0.678854	1.235472	1.220982	Russian Federation	14.35
48	1.266653	1.296742	0.615350	0.972800	0.919258	Germany	14.25

	IT.NET.USER.P2	NY.GNP.PCAP.PP.CD	SE.SEC.ENRR	SE.TER.ENRR	SP.POP.1524.TO.UN	Pays	Score
15	2.184787	1.702272	1.175855	1.950842	0.260830	Korea, Rep.	20.29
33	1.447463	2.239240	1.035853	1.762995	1.331004	United States	20.08
2	0.641004	0.538406	1.259486	0.681782	1.192399	Brazil	12.58
4	0.235485	0.433473	0.706352	0.402809	1.918316	China	11.50
5	0.519118	0.326610	1.091381	0.862176	0.472196	Colombia	9.28
32	0.163468	-0.820426	1.214275	1.670500	0.066830	Ukraine	8.47
13	-0.017903	0.775908	0.236952	1.378093	0.715522	Iran, Islamic Rep.	7.44
19	0.595935	0.728257	0.357749	-0.210547	0.938671	Mexico	6.87
25	0.407826	-0.538515	0.170730	0.072569	0.910227	Philippines	6.20
27	1.598854	1.507631	0.532078	0.765949	-1.627177	Slovak Republic	4.96

	NET.USER	ECO	SEC	TER	POP	Score
Income Group
High income: OECD	1.129015	1.117205	1.082073	1.064266	-0.095706	10.698667
High income: nonOECD	0.960337	1.146841	0.568577	0.610445	-0.950219	4.050667
Low income	-1.420378	-1.423266	-1.336054	-1.302546	0.373721	-12.287407
Lower middle income	-0.498394	-0.538922	-0.387760	-0.471286	0.124844	-4.221429
Upper middle income	0.227833	0.199142	0.262248	0.320579	0.080742	2.962727